Skip to content

Instantly share code, notes, and snippets.

@galvanic

galvanic/1608030254-df.txt

Last active September 8, 2016 07:55
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save galvanic/d76501e02a902ed623c8e0ceb4ac645b to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save galvanic/d76501e02a902ed623c8e0ceb4ac645b to your computer and use it in GitHub Desktop.
Download ZIP
Raw
1608030254-df.txt
metrics AUC FNR FPR error_test error_train
repetition dataset classifier attack % poisoned
1 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.50 1.00 0.00 0.66 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.66 0.83
empty 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.96 0.00 0.08 0.03 0.03
0.2 0.95 0.00 0.11 0.04 0.03
0.5 0.90 0.00 0.19 0.07 0.03
ham 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.01 0.09 0.03 0.03
0.5 0.50 1.00 0.00 0.66 0.83
focussed 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.96 0.00 0.07 0.03 0.02
0.2 0.96 0.00 0.07 0.03 0.02
0.5 0.96 0.00 0.07 0.03 0.01
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.00 0.02 0.01 0.00
0.2 0.99 0.00 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
empty 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.01 0.01 0.00
ham 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
focussed 0.0 0.99 0.00 0.02 0.01 0.00
0.1 0.99 0.00 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.53 0.92 0.01 0.62 0.56
0.2 0.50 1.00 0.00 0.66 0.53
0.5 0.50 1.00 0.00 0.67 0.33
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.05 0.05 0.05 0.04
0.2 0.95 0.06 0.05 0.06 0.04
0.5 0.92 0.10 0.06 0.09 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.06 0.06 0.05
0.2 0.92 0.09 0.06 0.08 0.06
0.5 0.78 0.42 0.02 0.29 0.14
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.96 0.07 0.02 0.05 0.05
0.2 0.94 0.11 0.02 0.08 0.06
0.5 0.92 0.14 0.02 0.10 0.05
trec2007-1607252257 adaline dictionary 0.0 0.95 0.01 0.09 0.03 0.03
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.66 0.74
0.5 0.50 1.00 0.00 0.66 0.84
empty 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.10 0.04 0.04
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.89 0.01 0.21 0.08 0.04
ham 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.91 0.01 0.17 0.06 0.05
0.2 0.90 0.01 0.19 0.07 0.05
0.5 0.89 0.01 0.21 0.08 0.04
focussed 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.09 0.04 0.03
0.2 0.94 0.01 0.11 0.04 0.03
0.5 0.95 0.01 0.09 0.04 0.02
logistic regression dictionary 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.02 0.01 0.01
0.5 0.98 0.01 0.02 0.02 0.00
empty 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.02 0.02 0.02 0.00
ham 0.0 0.98 0.02 0.02 0.02 0.01
0.1 0.98 0.01 0.03 0.01 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.97 0.04 0.02 0.03 0.01
focussed 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.01 0.01
0.5 0.98 0.02 0.02 0.02 0.00
naive bayes dictionary 0.0 0.94 0.10 0.03 0.07 0.07
0.1 0.53 0.88 0.07 0.61 0.55
0.2 0.52 0.93 0.03 0.63 0.50
0.5 0.53 0.93 0.01 0.63 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.10 0.05 0.08 0.07
0.2 0.92 0.11 0.06 0.09 0.07
0.5 0.87 0.18 0.08 0.15 0.08
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.09 0.08
0.2 0.90 0.12 0.08 0.11 0.08
0.5 0.84 0.22 0.10 0.18 0.09
focussed 0.0 0.94 0.10 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.09 0.08
0.2 0.74 0.51 0.01 0.34 0.28
0.5 0.81 0.33 0.06 0.24 0.12
trec2007-1607252259 adaline dictionary 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.88 0.01 0.24 0.09 0.08
0.2 0.87 0.01 0.25 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.90 0.01 0.19 0.07 0.07
0.2 0.89 0.01 0.22 0.08 0.06
0.5 0.79 0.01 0.41 0.14 0.07
ham 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.89 0.02 0.20 0.08 0.07
0.2 0.83 0.02 0.32 0.12 0.09
0.5 0.76 0.02 0.47 0.17 0.08
focussed 0.0 0.92 0.01 0.16 0.06 0.06
0.1 0.89 0.01 0.21 0.08 0.07
0.2 0.90 0.01 0.18 0.07 0.06
0.5 0.82 0.02 0.34 0.13 0.06
logistic regression dictionary 0.0 0.95 0.02 0.08 0.04 0.04
0.1 0.94 0.03 0.08 0.05 0.04
0.2 0.94 0.06 0.06 0.06 0.05
0.5 0.94 0.04 0.07 0.05 0.03
empty 0.0 0.92 0.01 0.15 0.05 0.05
0.1 0.95 0.03 0.07 0.04 0.03
0.2 0.95 0.02 0.08 0.04 0.03
0.5 0.95 0.05 0.05 0.05 0.02
ham 0.0 0.95 0.03 0.08 0.04 0.04
0.1 0.94 0.02 0.10 0.04 0.04
0.2 0.95 0.03 0.06 0.04 0.03
0.5 0.94 0.03 0.10 0.05 0.02
focussed 0.0 0.95 0.04 0.05 0.05 0.05
0.1 0.93 0.01 0.14 0.05 0.05
0.2 0.95 0.02 0.08 0.04 0.03
0.5 0.95 0.03 0.07 0.05 0.02
naive bayes dictionary 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.62 0.65 0.11 0.47 0.41
0.2 0.58 0.73 0.10 0.52 0.42
0.5 0.52 0.87 0.09 0.61 0.31
empty 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.88 0.04 0.19 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.81 0.22 0.17 0.20 0.10
ham 0.0 0.90 0.05 0.16 0.08 0.08
0.1 0.87 0.04 0.22 0.10 0.09
0.2 0.87 0.05 0.22 0.11 0.08
0.5 0.78 0.24 0.20 0.23 0.11
focussed 0.0 0.90 0.05 0.16 0.08 0.08
0.1 0.86 0.03 0.25 0.11 0.10
0.2 0.89 0.05 0.17 0.09 0.08
0.5 0.58 0.75 0.09 0.53 0.26
2 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.66 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.10 0.04 0.03
0.5 0.91 0.00 0.19 0.06 0.03
ham 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.10 0.04 0.03
0.5 0.50 1.00 0.00 0.67 0.83
focussed 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.96 0.00 0.07 0.03 0.02
0.5 0.96 0.00 0.09 0.03 0.02
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.00 0.02 0.01 0.00
0.5 0.99 0.01 0.01 0.01 0.00
empty 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.00 0.02 0.01 0.00
0.5 0.99 0.01 0.01 0.01 0.00
ham 0.0 0.99 0.00 0.02 0.01 0.00
0.1 0.99 0.00 0.02 0.01 0.00
0.2 0.99 0.00 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
focussed 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.53 0.92 0.01 0.62 0.55
0.2 0.50 1.00 0.00 0.66 0.53
0.5 0.50 1.00 0.00 0.67 0.33
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.05 0.04 0.05 0.04
0.2 0.95 0.06 0.05 0.06 0.04
0.5 0.92 0.10 0.07 0.09 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.06 0.06 0.05
0.2 0.93 0.09 0.06 0.08 0.06
0.5 0.77 0.45 0.01 0.30 0.15
focussed 0.0 0.96 0.05 0.02 0.04 0.04
0.1 0.95 0.05 0.04 0.05 0.04
0.2 0.93 0.08 0.06 0.07 0.06
0.5 0.78 0.42 0.01 0.28 0.14
trec2007-1607252257 adaline dictionary 0.0 0.95 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.66 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.94 0.01 0.10 0.04 0.04
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.89 0.01 0.21 0.08 0.04
ham 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.91 0.01 0.17 0.06 0.05
0.2 0.90 0.01 0.19 0.07 0.05
0.5 0.89 0.01 0.22 0.08 0.04
focussed 0.0 0.95 0.01 0.08 0.03 0.03
0.1 0.94 0.01 0.11 0.04 0.04
0.2 0.95 0.01 0.09 0.03 0.03
0.5 0.92 0.01 0.14 0.05 0.03
logistic regression dictionary 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
empty 0.0 0.98 0.01 0.03 0.01 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.04 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
ham 0.0 0.98 0.01 0.02 0.01 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
focussed 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.01 0.02 0.01 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.02 0.03 0.02 0.00
naive bayes dictionary 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.53 0.88 0.07 0.61 0.54
0.2 0.52 0.93 0.03 0.62 0.51
0.5 0.53 0.93 0.01 0.62 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.10 0.05 0.08 0.07
0.2 0.92 0.11 0.06 0.09 0.07
0.5 0.87 0.19 0.08 0.15 0.08
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.09 0.08
0.2 0.90 0.12 0.08 0.11 0.08
0.5 0.84 0.23 0.10 0.19 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.08 0.08
0.2 0.90 0.18 0.03 0.13 0.10
0.5 0.53 0.90 0.04 0.61 0.30
trec2007-1607252259 adaline dictionary 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.01 0.22 0.08 0.08
0.2 0.87 0.01 0.25 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.01 0.16 0.06 0.06
0.1 0.91 0.01 0.18 0.07 0.06
0.2 0.88 0.01 0.23 0.08 0.07
0.5 0.79 0.01 0.41 0.14 0.07
ham 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.87 0.01 0.24 0.09 0.08
0.2 0.83 0.02 0.33 0.12 0.09
0.5 0.75 0.02 0.47 0.17 0.08
focussed 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.87 0.02 0.24 0.10 0.08
0.2 0.88 0.01 0.22 0.08 0.06
0.5 0.88 0.01 0.23 0.09 0.04
logistic regression dictionary 0.0 0.95 0.06 0.04 0.06 0.06
0.1 0.95 0.04 0.07 0.05 0.05
0.2 0.93 0.05 0.08 0.06 0.05
0.5 0.94 0.03 0.10 0.05 0.02
empty 0.0 0.91 0.01 0.17 0.06 0.06
0.1 0.94 0.02 0.09 0.04 0.04
0.2 0.95 0.03 0.06 0.04 0.03
0.5 0.95 0.05 0.05 0.05 0.03
ham 0.0 0.95 0.05 0.04 0.05 0.05
0.1 0.95 0.02 0.07 0.04 0.03
0.2 0.96 0.03 0.05 0.04 0.03
0.5 0.94 0.02 0.09 0.04 0.02
focussed 0.0 0.93 0.01 0.12 0.05 0.05
0.1 0.94 0.03 0.08 0.05 0.14
0.2 0.95 0.03 0.06 0.04 0.03
0.5 0.95 0.04 0.07 0.05 0.02
naive bayes dictionary 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.63 0.63 0.11 0.46 0.42
0.2 0.58 0.73 0.11 0.52 0.42
0.5 0.52 0.88 0.09 0.61 0.31
empty 0.0 0.90 0.04 0.17 0.08 0.08
0.1 0.89 0.04 0.18 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.80 0.23 0.17 0.21 0.10
ham 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.04 0.22 0.10 0.09
0.2 0.87 0.05 0.21 0.11 0.08
0.5 0.79 0.22 0.20 0.21 0.11
focussed 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.83 0.04 0.31 0.13 0.12
0.2 0.85 0.08 0.21 0.12 0.10
0.5 0.44 0.89 0.22 0.67 0.34
3 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.00 0.07 0.02 0.03
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.94 0.00 0.11 0.04 0.03
0.5 0.90 0.00 0.19 0.07 0.03
ham 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.10 0.03 0.03
0.5 0.50 1.00 0.00 0.66 0.83
focussed 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.96 0.00 0.08 0.03 0.02
0.2 0.50 1.00 0.00 0.66 0.73
0.5 0.95 0.00 0.10 0.04 0.02
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.00 0.02 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.98 0.02 0.01 0.02 0.00
empty 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.00 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.00 0.02 0.01 0.00
ham 0.0 0.99 0.00 0.02 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
focussed 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.01 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.53 0.92 0.01 0.62 0.55
0.2 0.50 1.00 0.00 0.66 0.53
0.5 0.50 1.00 0.00 0.67 0.33
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.05 0.04 0.05 0.04
0.2 0.95 0.06 0.05 0.06 0.04
0.5 0.92 0.10 0.06 0.09 0.04
ham 0.0 0.96 0.05 0.02 0.04 0.04
0.1 0.94 0.06 0.06 0.06 0.05
0.2 0.93 0.09 0.06 0.08 0.06
0.5 0.77 0.44 0.01 0.30 0.15
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.96 0.06 0.03 0.05 0.04
0.2 0.94 0.06 0.06 0.06 0.05
0.5 0.76 0.48 0.01 0.32 0.16
trec2007-1607252257 adaline dictionary 0.0 0.95 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.94 0.01 0.10 0.04 0.04
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.89 0.01 0.21 0.08 0.04
ham 0.0 0.95 0.01 0.09 0.03 0.03
0.1 0.92 0.01 0.16 0.06 0.05
0.2 0.90 0.01 0.18 0.07 0.05
0.5 0.88 0.01 0.22 0.08 0.04
focussed 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.09 0.04 0.03
0.2 0.94 0.01 0.12 0.04 0.04
0.5 0.50 1.00 0.00 0.67 0.83
logistic regression dictionary 0.0 0.98 0.02 0.02 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
empty 0.0 0.98 0.01 0.03 0.01 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.02 0.02 0.00
ham 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
focussed 0.0 0.98 0.01 0.03 0.01 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.97 0.01 0.04 0.02 0.01
naive bayes dictionary 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.53 0.87 0.08 0.61 0.54
0.2 0.52 0.93 0.04 0.63 0.50
0.5 0.53 0.94 0.01 0.62 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.09 0.04 0.08 0.07
0.2 0.92 0.11 0.06 0.09 0.07
0.5 0.87 0.19 0.08 0.15 0.07
ham 0.0 0.94 0.10 0.03 0.07 0.07
0.1 0.92 0.09 0.06 0.08 0.07
0.2 0.90 0.12 0.08 0.11 0.08
0.5 0.84 0.23 0.10 0.19 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.05 0.09 0.08
0.2 0.90 0.16 0.03 0.12 0.10
0.5 0.58 0.82 0.03 0.55 0.28
trec2007-1607252259 adaline dictionary 0.0 0.92 0.01 0.14 0.06 0.06
0.1 0.88 0.01 0.23 0.09 0.08
0.2 0.87 0.01 0.25 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.01 0.14 0.06 0.06
0.1 0.91 0.01 0.17 0.06 0.06
0.2 0.89 0.01 0.21 0.08 0.06
0.5 0.80 0.01 0.40 0.14 0.07
ham 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.01 0.22 0.08 0.07
0.2 0.84 0.02 0.30 0.11 0.09
0.5 0.76 0.02 0.46 0.17 0.08
focussed 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.84 0.01 0.31 0.11 0.09
0.2 0.84 0.01 0.32 0.12 0.09
0.5 0.85 0.01 0.29 0.10 0.05
logistic regression dictionary 0.0 0.95 0.06 0.04 0.06 0.05
0.1 0.92 0.02 0.14 0.06 0.05
0.2 0.93 0.04 0.10 0.06 0.04
0.5 0.93 0.04 0.10 0.06 0.03
empty 0.0 0.95 0.02 0.08 0.04 0.04
0.1 0.95 0.03 0.07 0.04 0.04
0.2 0.95 0.03 0.07 0.04 0.03
0.5 0.95 0.02 0.08 0.04 0.02
ham 0.0 0.94 0.02 0.09 0.04 0.04
0.1 0.95 0.03 0.07 0.04 0.04
0.2 0.93 0.02 0.13 0.05 0.04
0.5 0.94 0.02 0.09 0.05 0.02
focussed 0.0 0.94 0.02 0.09 0.04 0.04
0.1 0.90 0.02 0.17 0.07 0.06
0.2 0.95 0.04 0.06 0.04 0.03
0.5 0.95 0.02 0.07 0.04 0.02
naive bayes dictionary 0.0 0.90 0.04 0.17 0.08 0.08
0.1 0.62 0.65 0.11 0.47 0.42
0.2 0.58 0.73 0.10 0.52 0.42
0.5 0.51 0.88 0.10 0.61 0.31
empty 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.89 0.04 0.18 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.81 0.22 0.16 0.20 0.10
ham 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.04 0.22 0.10 0.09
0.2 0.87 0.05 0.20 0.10 0.08
0.5 0.78 0.24 0.20 0.22 0.11
focussed 0.0 0.89 0.04 0.17 0.08 0.08
0.1 0.90 0.05 0.15 0.08 0.17
0.2 0.85 0.04 0.26 0.12 0.09
0.5 0.74 0.35 0.16 0.29 0.14
4 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.96 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.11 0.04 0.03
0.5 0.90 0.00 0.20 0.07 0.03
ham 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.95 0.00 0.10 0.03 0.03
0.2 0.95 0.00 0.09 0.03 0.03
0.5 0.50 1.00 0.00 0.67 0.83
focussed 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.96 0.00 0.07 0.03 0.02
0.2 0.96 0.00 0.07 0.03 0.02
0.5 0.50 1.00 0.00 0.67 0.83
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
empty 0.0 0.99 0.00 0.01 0.01 0.00
0.1 0.99 0.01 0.02 0.01 0.00
0.2 0.99 0.00 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
ham 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.02 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.01 0.01 0.00
focussed 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.02 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.02 0.04 0.04
0.1 0.54 0.92 0.01 0.61 0.56
0.2 0.50 1.00 0.00 0.66 0.53
0.5 0.50 1.00 0.00 0.67 0.33
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.05 0.04 0.05 0.04
0.2 0.95 0.06 0.05 0.05 0.04
0.5 0.92 0.10 0.06 0.09 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.06 0.06 0.05
0.2 0.93 0.09 0.06 0.08 0.06
0.5 0.77 0.44 0.01 0.30 0.15
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.84 0.32 0.00 0.21 0.19
0.2 0.93 0.09 0.04 0.07 0.06
0.5 0.89 0.17 0.05 0.13 0.07
trec2007-1607252257 adaline dictionary 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.66 0.70
0.2 0.50 1.00 0.00 0.66 0.73
0.5 0.50 1.00 0.00 0.66 0.83
empty 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.10 0.04 0.03
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.89 0.01 0.21 0.07 0.04
ham 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.91 0.01 0.16 0.06 0.05
0.2 0.91 0.01 0.17 0.06 0.06
0.5 0.89 0.01 0.21 0.08 0.04
focussed 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.10 0.04 0.04
0.2 0.93 0.01 0.12 0.05 0.03
0.5 0.95 0.01 0.09 0.04 0.02
logistic regression dictionary 0.0 0.98 0.01 0.02 0.01 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.01 0.02 0.02 0.00
empty 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.03 0.01 0.00
ham 0.0 0.98 0.02 0.02 0.02 0.02
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.01 0.02 0.01 0.01
0.5 0.98 0.01 0.04 0.02 0.00
focussed 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
naive bayes dictionary 0.0 0.94 0.09 0.04 0.07 0.07
0.1 0.54 0.86 0.06 0.59 0.53
0.2 0.52 0.93 0.04 0.63 0.50
0.5 0.53 0.93 0.01 0.63 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.10 0.05 0.08 0.07
0.2 0.91 0.11 0.06 0.09 0.07
0.5 0.87 0.19 0.08 0.15 0.08
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.08 0.08
0.2 0.90 0.12 0.08 0.10 0.08
0.5 0.84 0.23 0.09 0.18 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.11 0.03 0.09 0.08
0.2 0.89 0.18 0.04 0.13 0.10
0.5 0.84 0.29 0.04 0.21 0.11
trec2007-1607252259 adaline dictionary 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.01 0.23 0.08 0.08
0.2 0.87 0.01 0.25 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.02 0.14 0.06 0.06
0.1 0.90 0.01 0.18 0.07 0.06
0.2 0.89 0.01 0.21 0.08 0.07
0.5 0.79 0.01 0.41 0.14 0.07
ham 0.0 0.92 0.02 0.14 0.06 0.06
0.1 0.88 0.02 0.21 0.08 0.07
0.2 0.84 0.01 0.31 0.11 0.09
0.5 0.76 0.02 0.47 0.17 0.08
focussed 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.90 0.01 0.18 0.07 0.06
0.2 0.86 0.01 0.27 0.10 0.07
0.5 0.77 0.01 0.45 0.16 0.08
logistic regression dictionary 0.0 0.95 0.02 0.09 0.04 0.04
0.1 0.93 0.03 0.12 0.06 0.06
0.2 0.91 0.02 0.17 0.07 0.06
0.5 0.93 0.03 0.11 0.05 0.03
empty 0.0 0.96 0.03 0.06 0.04 0.04
0.1 0.96 0.04 0.05 0.04 0.04
0.2 0.94 0.01 0.11 0.05 0.04
0.5 0.90 0.01 0.19 0.07 0.04
ham 0.0 0.95 0.05 0.05 0.05 0.05
0.1 0.95 0.03 0.06 0.04 0.04
0.2 0.95 0.02 0.09 0.04 0.03
0.5 0.95 0.02 0.07 0.04 0.02
focussed 0.0 0.95 0.04 0.06 0.05 0.04
0.1 0.95 0.05 0.06 0.05 0.05
0.2 0.92 0.04 0.13 0.07 0.05
0.5 0.95 0.03 0.07 0.05 0.02
naive bayes dictionary 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.62 0.65 0.11 0.47 0.42
0.2 0.58 0.73 0.11 0.52 0.42
0.5 0.52 0.88 0.09 0.61 0.31
empty 0.0 0.90 0.04 0.17 0.08 0.08
0.1 0.89 0.04 0.17 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.81 0.22 0.17 0.20 0.10
ham 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.04 0.22 0.10 0.09
0.2 0.88 0.05 0.20 0.10 0.08
0.5 0.78 0.23 0.20 0.22 0.11
focussed 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.04 0.22 0.10 0.09
0.2 0.85 0.05 0.26 0.12 0.09
0.5 0.42 0.99 0.16 0.72 0.35
5 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.03 0.03
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.66 0.74
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.00 0.07 0.02 0.03
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.94 0.00 0.11 0.04 0.03
0.5 0.90 0.00 0.19 0.07 0.03
ham 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.09 0.03 0.02
0.5 0.50 1.00 0.00 0.67 0.83
focussed 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.96 0.00 0.07 0.03 0.02
0.5 0.50 1.00 0.00 0.67 0.83
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
empty 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.00 0.02 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.01 0.01 0.00
ham 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.02 0.01 0.00
0.2 0.99 0.00 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
focussed 0.0 0.99 0.00 0.01 0.01 0.00
0.1 0.99 0.00 0.02 0.01 0.00
0.2 0.99 0.00 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.54 0.92 0.01 0.62 0.55
0.2 0.50 0.99 0.00 0.66 0.53
0.5 0.50 1.00 0.00 0.67 0.33
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.05 0.04 0.05 0.04
0.2 0.95 0.06 0.05 0.06 0.04
0.5 0.92 0.10 0.06 0.09 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.06 0.06 0.05
0.2 0.92 0.09 0.06 0.08 0.06
0.5 0.77 0.45 0.01 0.30 0.15
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.96 0.05 0.04 0.05 0.04
0.2 0.95 0.09 0.01 0.07 0.06
0.5 0.62 0.76 0.01 0.51 0.26
trec2007-1607252257 adaline dictionary 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.66 0.83
empty 0.0 0.95 0.01 0.09 0.03 0.03
0.1 0.94 0.01 0.11 0.04 0.04
0.2 0.93 0.01 0.12 0.05 0.04
0.5 0.89 0.01 0.22 0.08 0.04
ham 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.91 0.01 0.17 0.06 0.05
0.2 0.90 0.01 0.19 0.07 0.05
0.5 0.89 0.01 0.21 0.08 0.04
focussed 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.94 0.01 0.11 0.04 0.04
0.2 0.94 0.01 0.11 0.04 0.03
0.5 0.94 0.01 0.10 0.04 0.02
logistic regression dictionary 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.02 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
empty 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.02 0.02 0.00
ham 0.0 0.98 0.01 0.03 0.01 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.01 0.04 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
focussed 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
naive bayes dictionary 0.0 0.94 0.10 0.03 0.07 0.07
0.1 0.54 0.86 0.06 0.59 0.54
0.2 0.52 0.93 0.04 0.63 0.51
0.5 0.53 0.93 0.01 0.62 0.31
empty 0.0 0.94 0.08 0.03 0.07 0.07
0.1 0.93 0.10 0.05 0.08 0.07
0.2 0.91 0.11 0.06 0.09 0.07
0.5 0.87 0.18 0.08 0.15 0.07
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.09 0.08
0.2 0.90 0.12 0.08 0.11 0.08
0.5 0.84 0.23 0.09 0.18 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.11 0.02 0.08 0.08
0.2 0.88 0.21 0.03 0.15 0.12
0.5 0.56 0.88 0.01 0.59 0.29
trec2007-1607252259 adaline dictionary 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.01 0.23 0.08 0.08
0.2 0.87 0.01 0.26 0.09 0.07
0.5 0.50 1.00 0.00 0.66 0.84
empty 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.91 0.01 0.17 0.07 0.06
0.2 0.89 0.01 0.21 0.08 0.06
0.5 0.80 0.01 0.40 0.14 0.07
ham 0.0 0.92 0.02 0.14 0.06 0.06
0.1 0.88 0.02 0.22 0.08 0.07
0.2 0.84 0.02 0.31 0.11 0.09
0.5 0.76 0.02 0.46 0.16 0.09
focussed 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.81 0.01 0.37 0.13 0.12
0.2 0.86 0.01 0.27 0.09 0.08
0.5 0.88 0.01 0.23 0.09 0.04
logistic regression dictionary 0.0 0.95 0.02 0.08 0.04 0.04
0.1 0.91 0.02 0.17 0.07 0.06
0.2 0.94 0.06 0.07 0.06 0.05
0.5 0.93 0.06 0.07 0.07 0.03
empty 0.0 0.95 0.03 0.06 0.04 0.04
0.1 0.93 0.01 0.13 0.05 0.04
0.2 0.95 0.03 0.07 0.04 0.03
0.5 0.93 0.01 0.12 0.05 0.02
ham 0.0 0.95 0.04 0.06 0.04 0.04
0.1 0.94 0.01 0.11 0.04 0.04
0.2 0.95 0.02 0.08 0.04 0.04
0.5 0.95 0.03 0.06 0.04 0.02
focussed 0.0 0.95 0.04 0.05 0.05 0.05
0.1 0.94 0.04 0.07 0.05 0.04
0.2 0.95 0.04 0.06 0.05 0.04
0.5 0.95 0.03 0.08 0.04 0.02
naive bayes dictionary 0.0 0.90 0.05 0.16 0.08 0.08
0.1 0.62 0.64 0.11 0.46 0.42
0.2 0.58 0.73 0.11 0.52 0.42
0.5 0.51 0.88 0.10 0.61 0.31
empty 0.0 0.90 0.04 0.17 0.08 0.08
0.1 0.89 0.04 0.17 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.81 0.22 0.16 0.20 0.10
ham 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.04 0.22 0.10 0.09
0.2 0.87 0.05 0.21 0.10 0.08
0.5 0.79 0.22 0.20 0.22 0.11
focussed 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.04 0.23 0.10 0.19
0.2 0.85 0.05 0.25 0.12 0.10
0.5 0.39 0.98 0.23 0.73 0.36
6 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.50 1.00 0.00 0.66 0.70
0.2 0.50 1.00 0.00 0.66 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.97 0.00 0.07 0.02 0.02
0.1 0.96 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.11 0.04 0.03
0.5 0.90 0.00 0.19 0.07 0.03
ham 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.09 0.03 0.03
0.5 0.50 1.00 0.00 0.67 0.83
focussed 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.96 0.00 0.08 0.03 0.02
0.2 0.96 0.00 0.08 0.03 0.02
0.5 0.50 1.00 0.00 0.67 0.83
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
empty 0.0 0.99 0.00 0.02 0.01 0.00
0.1 0.99 0.00 0.02 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.00 0.02 0.01 0.00
ham 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.00 0.02 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.98 0.01 0.02 0.01 0.00
focussed 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.54 0.92 0.01 0.61 0.55
0.2 0.50 1.00 0.00 0.66 0.53
0.5 0.50 1.00 0.00 0.67 0.33
empty 0.0 0.96 0.05 0.02 0.04 0.04
0.1 0.95 0.05 0.04 0.05 0.04
0.2 0.94 0.06 0.05 0.06 0.04
0.5 0.93 0.09 0.06 0.08 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.05 0.06 0.05
0.2 0.93 0.09 0.06 0.08 0.06
0.5 0.78 0.43 0.02 0.29 0.14
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.05 0.06 0.05
0.2 0.85 0.28 0.02 0.19 0.16
0.5 0.76 0.48 0.00 0.32 0.16
trec2007-1607252257 adaline dictionary 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.94 0.01 0.10 0.04 0.03
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.89 0.01 0.22 0.08 0.04
ham 0.0 0.95 0.01 0.09 0.03 0.03
0.1 0.92 0.01 0.16 0.06 0.05
0.2 0.90 0.01 0.19 0.07 0.05
0.5 0.89 0.01 0.22 0.08 0.04
focussed 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.10 0.04 0.03
0.2 0.95 0.01 0.10 0.04 0.03
0.5 0.95 0.01 0.10 0.04 0.02
logistic regression dictionary 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.02 0.02 0.02 0.01
0.5 0.97 0.03 0.02 0.03 0.01
empty 0.0 0.98 0.01 0.02 0.01 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
ham 0.0 0.97 0.01 0.04 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.03 0.01 0.01
0.5 0.98 0.02 0.02 0.02 0.00
focussed 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.02 0.03 0.02 0.01
0.2 0.98 0.02 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
naive bayes dictionary 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.53 0.87 0.08 0.61 0.55
0.2 0.52 0.93 0.04 0.63 0.51
0.5 0.53 0.93 0.01 0.62 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.10 0.04 0.08 0.07
0.2 0.91 0.11 0.06 0.10 0.07
0.5 0.87 0.18 0.08 0.15 0.07
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.11 0.06 0.09 0.08
0.2 0.90 0.12 0.08 0.11 0.08
0.5 0.84 0.22 0.09 0.18 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.10 0.04 0.08 0.07
0.2 0.91 0.13 0.05 0.11 0.08
0.5 0.55 0.90 0.00 0.60 0.30
trec2007-1607252259 adaline dictionary 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.88 0.01 0.23 0.08 0.08
0.2 0.87 0.01 0.25 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.91 0.01 0.17 0.07 0.06
0.2 0.90 0.01 0.20 0.07 0.06
0.5 0.79 0.01 0.40 0.14 0.07
ham 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.89 0.02 0.21 0.08 0.07
0.2 0.84 0.02 0.30 0.11 0.09
0.5 0.76 0.02 0.46 0.17 0.08
focussed 0.0 0.92 0.01 0.14 0.06 0.06
0.1 0.89 0.03 0.20 0.09 0.08
0.2 0.90 0.02 0.19 0.07 0.06
0.5 0.81 0.01 0.37 0.13 0.06
logistic regression dictionary 0.0 0.95 0.02 0.08 0.04 0.04
0.1 0.94 0.07 0.05 0.06 0.06
0.2 0.92 0.02 0.13 0.06 0.05
0.5 0.93 0.02 0.11 0.05 0.03
empty 0.0 0.95 0.04 0.05 0.04 0.04
0.1 0.95 0.03 0.06 0.04 0.04
0.2 0.94 0.01 0.11 0.05 0.04
0.5 0.95 0.04 0.06 0.05 0.02
ham 0.0 0.95 0.02 0.08 0.04 0.04
0.1 0.91 0.01 0.16 0.06 0.06
0.2 0.95 0.02 0.09 0.04 0.03
0.5 0.95 0.04 0.06 0.05 0.02
focussed 0.0 0.95 0.02 0.08 0.04 0.04
0.1 0.93 0.03 0.11 0.06 0.05
0.2 0.95 0.03 0.08 0.04 0.03
0.5 0.93 0.01 0.13 0.05 0.02
naive bayes dictionary 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.62 0.65 0.11 0.47 0.42
0.2 0.58 0.73 0.11 0.52 0.42
0.5 0.51 0.87 0.10 0.62 0.30
empty 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.89 0.04 0.18 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.81 0.22 0.17 0.20 0.10
ham 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.05 0.22 0.10 0.09
0.2 0.88 0.05 0.19 0.10 0.08
0.5 0.79 0.23 0.19 0.22 0.11
focussed 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.88 0.04 0.21 0.10 0.18
0.2 0.86 0.07 0.21 0.12 0.09
0.5 0.63 0.70 0.04 0.48 0.24
7 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.50 1.00 0.00 0.66 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.66 0.83
empty 0.0 0.97 0.00 0.07 0.02 0.03
0.1 0.96 0.00 0.09 0.03 0.03
0.2 0.94 0.00 0.11 0.04 0.03
0.5 0.90 0.00 0.19 0.07 0.03
ham 0.0 0.97 0.00 0.07 0.02 0.03
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.10 0.04 0.03
0.5 0.50 1.00 0.00 0.66 0.83
focussed 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.96 0.00 0.07 0.03 0.03
0.2 0.96 0.00 0.07 0.03 0.02
0.5 0.50 1.00 0.00 0.67 0.83
logistic regression dictionary 0.0 0.99 0.00 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
empty 0.0 0.99 0.00 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
ham 0.0 0.99 0.01 0.02 0.01 0.00
0.1 0.99 0.01 0.02 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
focussed 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.02 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.01
0.5 0.99 0.00 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.53 0.92 0.01 0.62 0.55
0.2 0.50 1.00 0.00 0.67 0.53
0.5 0.50 1.00 0.00 0.67 0.33
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.05 0.04 0.05 0.04
0.2 0.95 0.06 0.05 0.05 0.04
0.5 0.92 0.10 0.07 0.09 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.06 0.06 0.05
0.2 0.93 0.09 0.06 0.08 0.06
0.5 0.78 0.43 0.02 0.29 0.14
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.96 0.06 0.03 0.05 0.04
0.2 0.85 0.30 0.00 0.20 0.16
0.5 0.89 0.19 0.02 0.14 0.07
trec2007-1607252257 adaline dictionary 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.66 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.10 0.04 0.04
0.2 0.93 0.01 0.12 0.05 0.04
0.5 0.89 0.01 0.21 0.07 0.04
ham 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.92 0.01 0.16 0.06 0.05
0.2 0.90 0.01 0.19 0.07 0.05
0.5 0.88 0.01 0.22 0.08 0.04
focussed 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.09 0.04 0.03
0.2 0.95 0.01 0.09 0.04 0.03
0.5 0.94 0.01 0.12 0.05 0.02
logistic regression dictionary 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.01 0.01
0.5 0.98 0.02 0.02 0.02 0.00
empty 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.02 0.02 0.02 0.00
ham 0.0 0.98 0.01 0.03 0.01 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.02 0.02 0.02 0.00
focussed 0.0 0.98 0.01 0.02 0.01 0.01
0.1 0.98 0.02 0.03 0.02 0.01
0.2 0.98 0.01 0.04 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
naive bayes dictionary 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.54 0.86 0.06 0.59 0.53
0.2 0.52 0.93 0.04 0.63 0.50
0.5 0.53 0.94 0.01 0.63 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.09 0.05 0.08 0.07
0.2 0.91 0.11 0.06 0.09 0.08
0.5 0.87 0.19 0.08 0.15 0.08
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.09 0.08
0.2 0.90 0.11 0.08 0.10 0.08
0.5 0.85 0.22 0.09 0.17 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.91 0.12 0.06 0.10 0.09
0.2 0.88 0.16 0.09 0.13 0.11
0.5 0.70 0.58 0.01 0.39 0.19
trec2007-1607252259 adaline dictionary 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.01 0.22 0.08 0.08
0.2 0.87 0.01 0.25 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.90 0.01 0.18 0.07 0.06
0.2 0.89 0.01 0.22 0.08 0.06
0.5 0.79 0.01 0.40 0.14 0.07
ham 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.01 0.22 0.08 0.07
0.2 0.84 0.02 0.30 0.11 0.09
0.5 0.75 0.02 0.48 0.17 0.08
focussed 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.88 0.00 0.24 0.08 0.08
0.2 0.88 0.01 0.24 0.08 0.07
0.5 0.82 0.02 0.35 0.13 0.06
logistic regression dictionary 0.0 0.95 0.02 0.07 0.04 0.04
0.1 0.95 0.04 0.07 0.05 0.04
0.2 0.95 0.03 0.07 0.05 0.04
0.5 0.94 0.04 0.07 0.05 0.03
empty 0.0 0.92 0.01 0.14 0.05 0.05
0.1 0.95 0.03 0.07 0.05 0.04
0.2 0.94 0.02 0.09 0.05 0.03
0.5 0.95 0.03 0.06 0.04 0.02
ham 0.0 0.96 0.02 0.07 0.04 0.04
0.1 0.95 0.02 0.08 0.04 0.04
0.2 0.95 0.03 0.06 0.04 0.03
0.5 0.95 0.03 0.07 0.04 0.02
focussed 0.0 0.95 0.05 0.06 0.05 0.05
0.1 0.95 0.04 0.05 0.05 0.04
0.2 0.93 0.03 0.11 0.05 0.04
0.5 0.91 0.01 0.16 0.06 0.03
naive bayes dictionary 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.63 0.64 0.11 0.46 0.42
0.2 0.58 0.73 0.11 0.52 0.42
0.5 0.52 0.87 0.10 0.61 0.31
empty 0.0 0.90 0.05 0.16 0.08 0.08
0.1 0.89 0.04 0.17 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.80 0.23 0.17 0.21 0.10
ham 0.0 0.90 0.05 0.16 0.08 0.08
0.1 0.87 0.04 0.21 0.10 0.09
0.2 0.87 0.05 0.20 0.10 0.08
0.5 0.79 0.23 0.19 0.22 0.11
focussed 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.89 0.04 0.19 0.09 0.08
0.2 0.90 0.05 0.16 0.08 0.07
0.5 0.73 0.39 0.16 0.31 0.16
8 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.10 0.04 0.03
0.5 0.91 0.00 0.19 0.06 0.03
ham 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.94 0.00 0.11 0.04 0.03
0.5 0.50 1.00 0.00 0.66 0.83
focussed 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.96 0.00 0.09 0.03 0.03
0.2 0.96 0.00 0.07 0.03 0.02
0.5 0.95 0.00 0.09 0.03 0.02
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.00 0.02 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
empty 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.00 0.02 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
ham 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
focussed 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.00 0.02 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.53 0.92 0.01 0.62 0.55
0.2 0.50 1.00 0.00 0.66 0.53
0.5 0.50 1.00 0.00 0.67 0.33
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.05 0.04 0.05 0.04
0.2 0.95 0.06 0.05 0.05 0.04
0.5 0.92 0.10 0.07 0.09 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.05 0.06 0.05
0.2 0.92 0.09 0.06 0.08 0.06
0.5 0.77 0.43 0.02 0.29 0.15
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.96 0.05 0.03 0.05 0.04
0.2 0.64 0.70 0.01 0.47 0.38
0.5 0.91 0.16 0.02 0.11 0.06
trec2007-1607252257 adaline dictionary 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.66 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.94 0.01 0.11 0.04 0.03
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.89 0.01 0.21 0.08 0.04
ham 0.0 0.95 0.01 0.08 0.03 0.03
0.1 0.91 0.01 0.16 0.06 0.05
0.2 0.90 0.01 0.19 0.07 0.05
0.5 0.89 0.01 0.22 0.08 0.04
focussed 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.94 0.01 0.12 0.04 0.04
0.2 0.95 0.01 0.10 0.04 0.03
0.5 0.94 0.01 0.11 0.04 0.02
logistic regression dictionary 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.02 0.02 0.02 0.01
empty 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.01 0.01
0.5 0.98 0.01 0.03 0.02 0.00
ham 0.0 0.98 0.02 0.02 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.02 0.02 0.02 0.01
focussed 0.0 0.98 0.02 0.02 0.02 0.02
0.1 0.98 0.01 0.03 0.01 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
naive bayes dictionary 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.53 0.87 0.07 0.60 0.55
0.2 0.52 0.93 0.04 0.63 0.51
0.5 0.53 0.94 0.01 0.62 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.10 0.05 0.08 0.07
0.2 0.91 0.11 0.06 0.09 0.08
0.5 0.87 0.18 0.08 0.14 0.07
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.05 0.09 0.08
0.2 0.90 0.12 0.08 0.10 0.08
0.5 0.84 0.23 0.09 0.18 0.09
focussed 0.0 0.94 0.08 0.03 0.07 0.07
0.1 0.90 0.12 0.08 0.10 0.09
0.2 0.70 0.59 0.01 0.40 0.31
0.5 0.64 0.63 0.08 0.45 0.22
trec2007-1607252259 adaline dictionary 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.01 0.24 0.09 0.08
0.2 0.87 0.01 0.25 0.09 0.07
0.5 0.50 1.00 0.00 0.66 0.83
empty 0.0 0.92 0.02 0.14 0.06 0.06
0.1 0.91 0.01 0.17 0.07 0.06
0.2 0.89 0.01 0.21 0.08 0.06
0.5 0.79 0.01 0.41 0.14 0.07
ham 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.01 0.22 0.08 0.07
0.2 0.84 0.01 0.32 0.11 0.09
0.5 0.76 0.02 0.47 0.17 0.08
focussed 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.83 0.01 0.33 0.12 0.11
0.2 0.85 0.01 0.29 0.11 0.08
0.5 0.79 0.01 0.40 0.14 0.07
logistic regression dictionary 0.0 0.91 0.01 0.17 0.06 0.06
0.1 0.93 0.02 0.11 0.05 0.05
0.2 0.93 0.03 0.11 0.06 0.04
0.5 0.93 0.03 0.11 0.06 0.03
empty 0.0 0.93 0.01 0.13 0.05 0.05
0.1 0.96 0.03 0.05 0.04 0.03
0.2 0.96 0.03 0.06 0.04 0.03
0.5 0.95 0.04 0.06 0.05 0.02
ham 0.0 0.95 0.04 0.06 0.04 0.04
0.1 0.95 0.04 0.05 0.04 0.04
0.2 0.95 0.02 0.07 0.04 0.03
0.5 0.95 0.03 0.06 0.04 0.02
focussed 0.0 0.95 0.04 0.05 0.04 0.05
0.1 0.95 0.02 0.09 0.04 0.04
0.2 0.95 0.03 0.07 0.04 0.04
0.5 0.95 0.03 0.08 0.04 0.02
naive bayes dictionary 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.63 0.64 0.11 0.46 0.42
0.2 0.58 0.73 0.11 0.52 0.42
0.5 0.51 0.87 0.10 0.61 0.31
empty 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.89 0.04 0.18 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.81 0.22 0.17 0.20 0.10
ham 0.0 0.90 0.04 0.17 0.08 0.08
0.1 0.87 0.04 0.21 0.10 0.09
0.2 0.87 0.05 0.21 0.10 0.08
0.5 0.79 0.23 0.20 0.22 0.11
focussed 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.03 0.22 0.10 0.19
0.2 0.61 0.69 0.09 0.49 0.39
0.5 0.44 0.96 0.16 0.69 0.35
9 trec2007-1607201347 adaline dictionary 0.0 0.97 0.00 0.07 0.02 0.02
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.66 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.11 0.04 0.03
0.5 0.90 0.00 0.19 0.07 0.03
ham 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.96 0.00 0.08 0.03 0.03
0.2 0.95 0.00 0.10 0.03 0.03
0.5 0.50 1.00 0.00 0.67 0.83
focussed 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.96 0.00 0.09 0.03 0.03
0.2 0.96 0.00 0.07 0.02 0.02
0.5 0.50 1.00 0.00 0.67 0.83
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.00 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
empty 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
ham 0.0 0.99 0.00 0.02 0.01 0.00
0.1 0.99 0.01 0.02 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.98 0.01 0.02 0.01 0.00
focussed 0.0 0.99 0.01 0.02 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.53 0.92 0.01 0.61 0.56
0.2 0.50 1.00 0.00 0.67 0.53
0.5 0.50 1.00 0.00 0.67 0.33
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.96 0.05 0.04 0.05 0.04
0.2 0.94 0.06 0.06 0.06 0.04
0.5 0.92 0.10 0.06 0.09 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.06 0.05 0.06 0.05
0.2 0.93 0.09 0.05 0.08 0.06
0.5 0.77 0.43 0.02 0.29 0.15
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.72 0.55 0.01 0.37 0.33
0.2 0.95 0.06 0.04 0.05 0.05
0.5 0.90 0.18 0.01 0.12 0.06
trec2007-1607252257 adaline dictionary 0.0 0.95 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.66 0.70
0.2 0.50 1.00 0.00 0.66 0.73
0.5 0.50 1.00 0.00 0.66 0.84
empty 0.0 0.95 0.01 0.09 0.03 0.03
0.1 0.94 0.01 0.11 0.04 0.03
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.89 0.01 0.22 0.08 0.04
ham 0.0 0.95 0.01 0.08 0.03 0.03
0.1 0.92 0.01 0.16 0.06 0.06
0.2 0.90 0.01 0.19 0.07 0.05
0.5 0.89 0.01 0.21 0.08 0.04
focussed 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.09 0.03 0.03
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.94 0.01 0.10 0.04 0.02
logistic regression dictionary 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.02 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
empty 0.0 0.98 0.00 0.04 0.02 0.01
0.1 0.98 0.01 0.03 0.01 0.01
0.2 0.98 0.02 0.02 0.02 0.01
0.5 0.98 0.01 0.04 0.02 0.00
ham 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
focussed 0.0 0.98 0.01 0.02 0.01 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.97 0.01 0.04 0.02 0.01
naive bayes dictionary 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.53 0.88 0.07 0.61 0.54
0.2 0.52 0.93 0.03 0.63 0.51
0.5 0.53 0.94 0.01 0.63 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.09 0.04 0.08 0.07
0.2 0.91 0.11 0.06 0.09 0.07
0.5 0.87 0.18 0.08 0.15 0.08
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.09 0.08
0.2 0.90 0.12 0.08 0.10 0.08
0.5 0.84 0.23 0.10 0.18 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.11 0.04 0.09 0.08
0.2 0.68 0.62 0.02 0.42 0.34
0.5 0.72 0.52 0.05 0.36 0.19
trec2007-1607252259 adaline dictionary 0.0 0.92 0.02 0.14 0.06 0.06
0.1 0.88 0.01 0.23 0.08 0.08
0.2 0.87 0.01 0.25 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.02 0.14 0.06 0.06
0.1 0.91 0.01 0.17 0.07 0.06
0.2 0.89 0.01 0.21 0.08 0.06
0.5 0.79 0.01 0.40 0.14 0.07
ham 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.88 0.02 0.22 0.08 0.07
0.2 0.84 0.02 0.31 0.11 0.09
0.5 0.76 0.02 0.46 0.17 0.08
focussed 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.86 0.01 0.26 0.10 0.08
0.2 0.86 0.02 0.26 0.10 0.08
0.5 0.82 0.01 0.35 0.12 0.06
logistic regression dictionary 0.0 0.95 0.03 0.07 0.04 0.04
0.1 0.94 0.03 0.09 0.05 0.04
0.2 0.94 0.05 0.07 0.06 0.04
0.5 0.93 0.03 0.10 0.05 0.03
empty 0.0 0.94 0.02 0.10 0.05 0.05
0.1 0.95 0.05 0.05 0.05 0.04
0.2 0.95 0.04 0.06 0.05 0.04
0.5 0.95 0.02 0.09 0.04 0.02
ham 0.0 0.95 0.03 0.06 0.04 0.04
0.1 0.94 0.02 0.09 0.05 0.04
0.2 0.94 0.02 0.10 0.05 0.04
0.5 0.95 0.03 0.07 0.04 0.02
focussed 0.0 0.95 0.03 0.07 0.04 0.04
0.1 0.95 0.03 0.07 0.04 0.04
0.2 0.93 0.04 0.10 0.06 0.05
0.5 0.94 0.05 0.07 0.06 0.03
naive bayes dictionary 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.63 0.64 0.11 0.46 0.42
0.2 0.58 0.73 0.10 0.52 0.41
0.5 0.52 0.88 0.09 0.62 0.31
empty 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.89 0.04 0.18 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.80 0.22 0.17 0.20 0.10
ham 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.04 0.21 0.10 0.09
0.2 0.87 0.05 0.21 0.10 0.08
0.5 0.78 0.24 0.20 0.23 0.11
focussed 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.85 0.05 0.26 0.12 0.10
0.2 0.87 0.08 0.17 0.11 0.09
0.5 0.58 0.80 0.04 0.55 0.28
10 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.66 0.83
empty 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.96 0.00 0.08 0.03 0.03
0.2 0.95 0.00 0.10 0.04 0.03
0.5 0.90 0.00 0.19 0.07 0.03
ham 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.09 0.03 0.03
0.5 0.50 1.00 0.00 0.67 0.83
focussed 0.0 0.96 0.00 0.07 0.02 0.03
0.1 0.96 0.00 0.08 0.03 0.03
0.2 0.96 0.00 0.07 0.03 0.02
0.5 0.50 1.00 0.00 0.66 0.83
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.00 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
empty 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
ham 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
focussed 0.0 0.99 0.00 0.02 0.01 0.00
0.1 0.99 0.00 0.02 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.54 0.92 0.01 0.61 0.55
0.2 0.50 1.00 0.00 0.66 0.53
0.5 0.50 1.00 0.00 0.67 0.33
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.05 0.04 0.05 0.04
0.2 0.95 0.06 0.05 0.05 0.05
0.5 0.92 0.10 0.07 0.09 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.06 0.06 0.05
0.2 0.93 0.09 0.06 0.08 0.06
0.5 0.78 0.44 0.01 0.29 0.15
focussed 0.0 0.96 0.05 0.02 0.04 0.04
0.1 0.95 0.06 0.05 0.05 0.05
0.2 0.94 0.07 0.05 0.06 0.05
0.5 0.89 0.17 0.06 0.13 0.06
trec2007-1607252257 adaline dictionary 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.66 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.66 0.83
empty 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.94 0.01 0.11 0.04 0.03
0.2 0.93 0.01 0.12 0.05 0.04
0.5 0.89 0.01 0.21 0.08 0.04
ham 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.92 0.01 0.16 0.06 0.05
0.2 0.90 0.01 0.19 0.07 0.05
0.5 0.88 0.01 0.23 0.08 0.04
focussed 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.08 0.03 0.03
0.2 0.95 0.01 0.10 0.04 0.03
0.5 0.90 0.02 0.18 0.07 0.04
logistic regression dictionary 0.0 0.98 0.03 0.01 0.02 0.02
0.1 0.98 0.01 0.02 0.01 0.01
0.2 0.98 0.02 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
empty 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.03 0.02 0.02 0.02
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
ham 0.0 0.98 0.03 0.02 0.02 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
focussed 0.0 0.98 0.01 0.03 0.01 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.02 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
naive bayes dictionary 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.54 0.86 0.06 0.59 0.53
0.2 0.52 0.93 0.03 0.63 0.51
0.5 0.53 0.93 0.01 0.62 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.10 0.04 0.08 0.07
0.2 0.92 0.11 0.06 0.09 0.07
0.5 0.86 0.19 0.08 0.15 0.08
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.08 0.08
0.2 0.90 0.12 0.08 0.11 0.09
0.5 0.84 0.23 0.10 0.18 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.12 0.03 0.09 0.08
0.2 0.89 0.15 0.06 0.12 0.10
0.5 0.69 0.51 0.11 0.38 0.19
trec2007-1607252259 adaline dictionary 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.01 0.23 0.08 0.08
0.2 0.87 0.01 0.24 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.91 0.01 0.17 0.07 0.06
0.2 0.88 0.01 0.23 0.08 0.07
0.5 0.79 0.01 0.40 0.14 0.07
ham 0.0 0.92 0.02 0.14 0.06 0.06
0.1 0.89 0.02 0.21 0.08 0.07
0.2 0.84 0.01 0.31 0.11 0.09
0.5 0.76 0.02 0.46 0.17 0.08
focussed 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.88 0.01 0.22 0.08 0.07
0.2 0.85 0.01 0.28 0.10 0.08
0.5 0.83 0.02 0.32 0.12 0.06
logistic regression dictionary 0.0 0.95 0.02 0.09 0.04 0.04
0.1 0.95 0.04 0.06 0.05 0.04
0.2 0.94 0.03 0.10 0.05 0.04
0.5 0.95 0.03 0.08 0.05 0.02
empty 0.0 0.95 0.04 0.05 0.04 0.04
0.1 0.95 0.02 0.09 0.04 0.04
0.2 0.94 0.01 0.11 0.04 0.03
0.5 0.95 0.02 0.09 0.04 0.02
ham 0.0 0.95 0.05 0.05 0.05 0.05
0.1 0.95 0.03 0.06 0.04 0.04
0.2 0.93 0.01 0.12 0.05 0.04
0.5 0.95 0.02 0.08 0.04 0.02
focussed 0.0 0.96 0.03 0.06 0.04 0.04
0.1 0.95 0.05 0.05 0.05 0.05
0.2 0.95 0.02 0.07 0.04 0.03
0.5 0.95 0.03 0.07 0.04 0.02
naive bayes dictionary 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.62 0.65 0.11 0.47 0.42
0.2 0.58 0.73 0.11 0.52 0.42
0.5 0.51 0.88 0.10 0.62 0.30
empty 0.0 0.90 0.04 0.17 0.08 0.08
0.1 0.89 0.04 0.18 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.81 0.22 0.17 0.20 0.10
ham 0.0 0.90 0.04 0.16 0.08 0.09
0.1 0.87 0.04 0.21 0.10 0.09
0.2 0.87 0.05 0.21 0.10 0.08
0.5 0.78 0.24 0.20 0.23 0.11
focussed 0.0 0.90 0.05 0.16 0.08 0.08
0.1 0.90 0.03 0.17 0.08 0.07
0.2 0.85 0.05 0.25 0.11 0.09
0.5 0.71 0.48 0.09 0.35 0.18
11 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.50 1.00 0.00 0.66 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.96 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.10 0.04 0.03
0.5 0.90 0.00 0.19 0.07 0.03
ham 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.95 0.00 0.10 0.03 0.03
0.2 0.95 0.00 0.10 0.03 0.03
0.5 0.50 1.00 0.00 0.67 0.83
focussed 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.96 0.00 0.08 0.03 0.02
0.2 0.96 0.00 0.07 0.03 0.02
0.5 0.96 0.00 0.07 0.03 0.01
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
empty 0.0 0.99 0.01 0.02 0.01 0.00
0.1 0.99 0.00 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.00 0.02 0.01 0.00
ham 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.02 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
focussed 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.00 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.53 0.92 0.01 0.62 0.55
0.2 0.50 1.00 0.00 0.67 0.53
0.5 0.50 1.00 0.00 0.67 0.33
empty 0.0 0.96 0.05 0.02 0.04 0.04
0.1 0.95 0.05 0.04 0.05 0.04
0.2 0.95 0.06 0.05 0.06 0.04
0.5 0.92 0.10 0.06 0.08 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.05 0.06 0.05
0.2 0.93 0.09 0.05 0.08 0.06
0.5 0.78 0.43 0.02 0.29 0.14
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.81 0.38 0.01 0.26 0.23
0.2 0.95 0.08 0.03 0.06 0.05
0.5 0.92 0.14 0.03 0.10 0.05
trec2007-1607252257 adaline dictionary 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.66 0.84
empty 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.94 0.01 0.10 0.04 0.04
0.2 0.94 0.01 0.12 0.05 0.04
0.5 0.89 0.01 0.22 0.08 0.04
ham 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.91 0.01 0.17 0.06 0.05
0.2 0.90 0.01 0.18 0.07 0.05
0.5 0.88 0.01 0.23 0.08 0.04
focussed 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.09 0.03 0.03
0.2 0.95 0.01 0.09 0.03 0.03
0.5 0.95 0.01 0.09 0.04 0.02
logistic regression dictionary 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.02 0.02 0.02 0.01
0.5 0.98 0.02 0.02 0.02 0.00
empty 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.01 0.04 0.02 0.01
0.5 0.98 0.02 0.02 0.02 0.01
ham 0.0 0.98 0.01 0.02 0.01 0.01
0.1 0.98 0.01 0.03 0.01 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.01 0.04 0.02 0.00
focussed 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.02 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
naive bayes dictionary 0.0 0.94 0.09 0.03 0.07 0.08
0.1 0.53 0.88 0.07 0.60 0.55
0.2 0.52 0.93 0.03 0.63 0.50
0.5 0.53 0.93 0.01 0.62 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.09 0.04 0.08 0.07
0.2 0.92 0.11 0.06 0.09 0.07
0.5 0.87 0.19 0.08 0.15 0.07
ham 0.0 0.94 0.10 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.08 0.08
0.2 0.90 0.12 0.08 0.10 0.08
0.5 0.84 0.22 0.10 0.18 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.05 0.08 0.07
0.2 0.88 0.18 0.07 0.14 0.11
0.5 0.70 0.58 0.01 0.39 0.19
trec2007-1607252259 adaline dictionary 0.0 0.92 0.01 0.14 0.06 0.06
0.1 0.88 0.01 0.23 0.09 0.08
0.2 0.87 0.01 0.25 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.90 0.01 0.18 0.07 0.06
0.2 0.88 0.01 0.22 0.08 0.06
0.5 0.80 0.01 0.40 0.14 0.07
ham 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.89 0.01 0.21 0.08 0.07
0.2 0.84 0.02 0.31 0.11 0.09
0.5 0.76 0.02 0.47 0.17 0.09
focussed 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.90 0.02 0.17 0.07 0.06
0.2 0.87 0.01 0.24 0.09 0.07
0.5 0.84 0.01 0.30 0.11 0.05
logistic regression dictionary 0.0 0.95 0.02 0.07 0.04 0.04
0.1 0.94 0.06 0.06 0.06 0.05
0.2 0.93 0.02 0.12 0.06 0.04
0.5 0.94 0.03 0.10 0.05 0.02
empty 0.0 0.95 0.03 0.07 0.04 0.04
0.1 0.95 0.04 0.06 0.05 0.04
0.2 0.95 0.06 0.04 0.05 0.04
0.5 0.94 0.01 0.11 0.05 0.02
ham 0.0 0.95 0.04 0.05 0.05 0.04
0.1 0.93 0.01 0.12 0.05 0.04
0.2 0.95 0.05 0.05 0.05 0.04
0.5 0.95 0.03 0.07 0.04 0.02
focussed 0.0 0.95 0.03 0.06 0.04 0.04
0.1 0.91 0.02 0.15 0.07 0.06
0.2 0.95 0.03 0.08 0.04 0.03
0.5 0.93 0.02 0.12 0.05 0.02
naive bayes dictionary 0.0 0.90 0.05 0.16 0.08 0.08
0.1 0.62 0.65 0.11 0.47 0.41
0.2 0.58 0.73 0.10 0.52 0.42
0.5 0.52 0.87 0.10 0.61 0.31
empty 0.0 0.90 0.04 0.17 0.08 0.08
0.1 0.89 0.04 0.17 0.09 0.08
0.2 0.89 0.05 0.18 0.09 0.07
0.5 0.81 0.22 0.16 0.20 0.10
ham 0.0 0.90 0.04 0.17 0.08 0.08
0.1 0.87 0.04 0.22 0.10 0.09
0.2 0.87 0.05 0.21 0.10 0.08
0.5 0.79 0.23 0.20 0.22 0.11
focussed 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.90 0.04 0.16 0.08 0.07
0.2 0.84 0.09 0.23 0.13 0.10
0.5 0.69 0.41 0.22 0.34 0.17
12 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.11 0.04 0.03
0.5 0.90 0.00 0.19 0.07 0.03
ham 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.10 0.03 0.03
0.5 0.50 1.00 0.00 0.67 0.83
focussed 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.96 0.00 0.07 0.03 0.02
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
empty 0.0 0.99 0.01 0.02 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.00 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
ham 0.0 0.99 0.00 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.01 0.01 0.00
focussed 0.0 0.99 0.02 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.53 0.92 0.01 0.61 0.55
0.2 0.50 1.00 0.00 0.66 0.53
0.5 0.50 1.00 0.00 0.66 0.33
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.05 0.04 0.05 0.04
0.2 0.95 0.06 0.05 0.06 0.04
0.5 0.92 0.10 0.06 0.09 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.06 0.05 0.06 0.05
0.2 0.92 0.09 0.06 0.08 0.06
0.5 0.78 0.43 0.02 0.29 0.14
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.08 0.01 0.06 0.05
0.2 0.94 0.07 0.05 0.06 0.05
0.5 0.89 0.17 0.06 0.13 0.07
trec2007-1607252257 adaline dictionary 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.66 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.95 0.01 0.08 0.03 0.03
0.1 0.94 0.01 0.11 0.04 0.03
0.2 0.93 0.01 0.12 0.05 0.04
0.5 0.89 0.01 0.21 0.08 0.04
ham 0.0 0.95 0.01 0.08 0.03 0.03
0.1 0.92 0.01 0.16 0.06 0.05
0.2 0.91 0.01 0.18 0.07 0.05
0.5 0.88 0.01 0.22 0.08 0.04
focussed 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.10 0.04 0.03
0.2 0.94 0.01 0.10 0.04 0.03
0.5 0.93 0.01 0.14 0.05 0.02
logistic regression dictionary 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
empty 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.02 0.02 0.00
ham 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.00 0.04 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
focussed 0.0 0.98 0.02 0.02 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
naive bayes dictionary 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.54 0.86 0.06 0.59 0.53
0.2 0.52 0.93 0.04 0.63 0.50
0.5 0.53 0.94 0.01 0.62 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.10 0.05 0.08 0.07
0.2 0.91 0.11 0.06 0.09 0.08
0.5 0.87 0.19 0.07 0.15 0.08
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.08 0.08
0.2 0.90 0.12 0.08 0.11 0.08
0.5 0.84 0.23 0.10 0.18 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.05 0.09 0.07
0.2 0.91 0.15 0.04 0.11 0.09
0.5 0.83 0.30 0.03 0.21 0.11
trec2007-1607252259 adaline dictionary 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.01 0.23 0.08 0.08
0.2 0.87 0.01 0.24 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.91 0.01 0.17 0.06 0.06
0.2 0.88 0.01 0.22 0.08 0.07
0.5 0.79 0.01 0.41 0.14 0.07
ham 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.01 0.22 0.08 0.08
0.2 0.84 0.02 0.31 0.11 0.09
0.5 0.75 0.02 0.48 0.17 0.08
focussed 0.0 0.92 0.02 0.14 0.06 0.06
0.1 0.90 0.01 0.19 0.07 0.06
0.2 0.89 0.02 0.20 0.08 0.06
0.5 0.77 0.02 0.45 0.16 0.08
logistic regression dictionary 0.0 0.94 0.01 0.11 0.05 0.04
0.1 0.92 0.10 0.06 0.08 0.07
0.2 0.94 0.06 0.06 0.06 0.05
0.5 0.92 0.02 0.14 0.06 0.03
empty 0.0 0.94 0.01 0.10 0.04 0.04
0.1 0.95 0.02 0.09 0.04 0.03
0.2 0.95 0.03 0.07 0.04 0.03
0.5 0.95 0.02 0.07 0.04 0.02
ham 0.0 0.95 0.05 0.05 0.05 0.04
0.1 0.94 0.02 0.09 0.05 0.04
0.2 0.95 0.05 0.06 0.05 0.04
0.5 0.94 0.04 0.08 0.05 0.03
focussed 0.0 0.95 0.03 0.08 0.05 0.05
0.1 0.95 0.04 0.07 0.05 0.04
0.2 0.95 0.04 0.06 0.05 0.04
0.5 0.96 0.04 0.05 0.04 0.02
naive bayes dictionary 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.62 0.65 0.11 0.47 0.41
0.2 0.58 0.73 0.11 0.52 0.42
0.5 0.52 0.87 0.10 0.61 0.31
empty 0.0 0.90 0.04 0.15 0.08 0.08
0.1 0.89 0.04 0.17 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.80 0.23 0.17 0.21 0.10
ham 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.04 0.21 0.10 0.09
0.2 0.87 0.05 0.20 0.10 0.08
0.5 0.79 0.23 0.20 0.22 0.11
focussed 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.88 0.04 0.20 0.09 0.08
0.2 0.82 0.04 0.32 0.14 0.11
0.5 0.72 0.41 0.15 0.32 0.16
13 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.96 0.00 0.08 0.03 0.03
0.2 0.94 0.00 0.11 0.04 0.03
0.5 0.90 0.00 0.19 0.07 0.03
ham 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.10 0.04 0.03
0.5 0.50 1.00 0.00 0.67 0.83
focussed 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.96 0.00 0.07 0.03 0.02
0.2 0.96 0.00 0.07 0.03 0.02
0.5 0.96 0.00 0.08 0.03 0.01
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.01 0.01 0.00
empty 0.0 0.99 0.00 0.02 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
ham 0.0 0.99 0.00 0.02 0.01 0.00
0.1 0.99 0.01 0.02 0.01 0.00
0.2 0.99 0.00 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
focussed 0.0 0.99 0.00 0.02 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.53 0.92 0.01 0.62 0.55
0.2 0.50 1.00 0.00 0.66 0.53
0.5 0.50 1.00 0.00 0.67 0.33
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.05 0.04 0.05 0.04
0.2 0.95 0.06 0.05 0.05 0.04
0.5 0.92 0.10 0.06 0.09 0.04
ham 0.0 0.96 0.05 0.02 0.04 0.04
0.1 0.94 0.06 0.06 0.06 0.05
0.2 0.92 0.09 0.06 0.08 0.06
0.5 0.79 0.41 0.02 0.28 0.14
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.06 0.04 0.05 0.04
0.2 0.94 0.08 0.04 0.07 0.05
0.5 0.90 0.17 0.02 0.12 0.06
trec2007-1607252257 adaline dictionary 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.66 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.66 0.84
empty 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.94 0.01 0.10 0.04 0.04
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.89 0.01 0.21 0.08 0.04
ham 0.0 0.95 0.01 0.08 0.03 0.03
0.1 0.91 0.01 0.16 0.06 0.05
0.2 0.90 0.01 0.18 0.07 0.05
0.5 0.89 0.01 0.22 0.08 0.04
focussed 0.0 0.95 0.01 0.08 0.03 0.03
0.1 0.93 0.01 0.13 0.05 0.04
0.2 0.95 0.01 0.08 0.03 0.03
0.5 0.95 0.01 0.08 0.04 0.02
logistic regression dictionary 0.0 0.98 0.01 0.02 0.01 0.01
0.1 0.98 0.01 0.03 0.01 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
empty 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.01 0.02 0.01 0.01
0.2 0.98 0.02 0.02 0.02 0.01
0.5 0.98 0.02 0.02 0.02 0.01
ham 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.01 0.04 0.02 0.00
focussed 0.0 0.98 0.02 0.02 0.02 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
naive bayes dictionary 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.53 0.86 0.07 0.60 0.54
0.2 0.52 0.93 0.03 0.63 0.51
0.5 0.53 0.94 0.01 0.63 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.09 0.04 0.08 0.07
0.2 0.92 0.11 0.05 0.09 0.07
0.5 0.87 0.18 0.07 0.15 0.07
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.10 0.05 0.08 0.08
0.2 0.90 0.12 0.08 0.11 0.08
0.5 0.84 0.22 0.10 0.18 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.10 0.04 0.08 0.08
0.2 0.65 0.69 0.02 0.47 0.37
0.5 0.55 0.89 0.00 0.59 0.29
trec2007-1607252259 adaline dictionary 0.0 0.92 0.01 0.14 0.06 0.06
0.1 0.87 0.01 0.24 0.09 0.08
0.2 0.87 0.01 0.25 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.91 0.01 0.17 0.07 0.06
0.2 0.88 0.01 0.22 0.08 0.07
0.5 0.80 0.01 0.40 0.14 0.07
ham 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.02 0.22 0.08 0.07
0.2 0.84 0.02 0.31 0.11 0.09
0.5 0.76 0.02 0.46 0.16 0.08
focussed 0.0 0.91 0.01 0.16 0.06 0.06
0.1 0.89 0.01 0.22 0.08 0.07
0.2 0.88 0.01 0.24 0.09 0.07
0.5 0.84 0.00 0.32 0.11 0.05
logistic regression dictionary 0.0 0.94 0.08 0.04 0.07 0.07
0.1 0.94 0.05 0.06 0.05 0.05
0.2 0.94 0.04 0.07 0.05 0.04
0.5 0.88 0.01 0.23 0.08 0.04
empty 0.0 0.95 0.04 0.06 0.05 0.05
0.1 0.95 0.03 0.06 0.04 0.03
0.2 0.92 0.02 0.15 0.06 0.05
0.5 0.96 0.05 0.04 0.04 0.02
ham 0.0 0.95 0.02 0.08 0.04 0.04
0.1 0.94 0.01 0.12 0.05 0.04
0.2 0.95 0.03 0.07 0.04 0.03
0.5 0.95 0.03 0.06 0.04 0.02
focussed 0.0 0.95 0.02 0.09 0.04 0.04
0.1 0.94 0.03 0.09 0.05 0.04
0.2 0.94 0.05 0.06 0.06 0.04
0.5 0.95 0.05 0.05 0.05 0.02
naive bayes dictionary 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.63 0.64 0.11 0.46 0.42
0.2 0.58 0.73 0.11 0.52 0.42
0.5 0.52 0.87 0.10 0.61 0.31
empty 0.0 0.90 0.04 0.17 0.08 0.08
0.1 0.89 0.04 0.18 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.80 0.23 0.17 0.21 0.10
ham 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.04 0.21 0.10 0.09
0.2 0.87 0.05 0.21 0.10 0.08
0.5 0.79 0.23 0.20 0.22 0.11
focussed 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.82 0.03 0.33 0.13 0.22
0.2 0.81 0.05 0.32 0.14 0.11
0.5 0.73 0.46 0.07 0.33 0.17
14 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.50 1.00 0.00 0.66 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.96 0.00 0.09 0.03 0.03
0.2 0.94 0.00 0.11 0.04 0.03
0.5 0.90 0.00 0.19 0.07 0.03
ham 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.10 0.04 0.03
0.5 0.50 1.00 0.00 0.66 0.83
focussed 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.96 0.00 0.07 0.02 0.02
0.2 0.95 0.00 0.11 0.04 0.03
0.5 0.50 1.00 0.00 0.66 0.83
logistic regression dictionary 0.0 0.99 0.01 0.02 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
empty 0.0 0.99 0.00 0.02 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
ham 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
focussed 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.00 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.53 0.92 0.01 0.62 0.56
0.2 0.50 1.00 0.00 0.67 0.53
0.5 0.50 1.00 0.00 0.66 0.34
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.05 0.04 0.05 0.04
0.2 0.94 0.06 0.06 0.06 0.05
0.5 0.92 0.10 0.06 0.09 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.06 0.06 0.05
0.2 0.93 0.09 0.06 0.08 0.06
0.5 0.78 0.43 0.01 0.29 0.14
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.96 0.06 0.03 0.05 0.04
0.2 0.92 0.16 0.01 0.11 0.08
0.5 0.59 0.80 0.01 0.54 0.27
trec2007-1607252257 adaline dictionary 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.66 0.73
0.5 0.50 1.00 0.00 0.66 0.83
empty 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.94 0.01 0.11 0.04 0.04
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.89 0.01 0.21 0.07 0.04
ham 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.91 0.01 0.17 0.06 0.05
0.2 0.90 0.01 0.19 0.07 0.05
0.5 0.89 0.01 0.21 0.08 0.04
focussed 0.0 0.95 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.09 0.04 0.03
0.2 0.94 0.01 0.12 0.04 0.04
0.5 0.94 0.02 0.10 0.04 0.02
logistic regression dictionary 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.01 0.02 0.01 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.02 0.02 0.02 0.01
empty 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.01 0.03 0.01 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
ham 0.0 0.98 0.01 0.02 0.01 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.03 0.02 0.03 0.01
focussed 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.02 0.02 0.00
naive bayes dictionary 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.54 0.86 0.07 0.60 0.53
0.2 0.52 0.93 0.03 0.63 0.50
0.5 0.53 0.94 0.01 0.63 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.09 0.05 0.08 0.07
0.2 0.92 0.10 0.06 0.09 0.08
0.5 0.87 0.19 0.08 0.15 0.08
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.09 0.08
0.2 0.90 0.11 0.08 0.10 0.08
0.5 0.84 0.22 0.09 0.18 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.10 0.04 0.08 0.08
0.2 0.90 0.16 0.05 0.12 0.10
0.5 0.53 0.93 0.01 0.63 0.31
trec2007-1607252259 adaline dictionary 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.88 0.01 0.24 0.09 0.08
0.2 0.87 0.01 0.24 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.91 0.01 0.17 0.06 0.06
0.2 0.89 0.01 0.22 0.08 0.06
0.5 0.79 0.01 0.42 0.14 0.07
ham 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.89 0.02 0.21 0.08 0.07
0.2 0.84 0.02 0.30 0.11 0.09
0.5 0.76 0.02 0.46 0.16 0.08
focussed 0.0 0.91 0.01 0.16 0.06 0.06
0.1 0.85 0.01 0.28 0.10 0.09
0.2 0.84 0.00 0.32 0.11 0.09
0.5 0.85 0.00 0.31 0.10 0.05
logistic regression dictionary 0.0 0.94 0.02 0.11 0.05 0.05
0.1 0.92 0.02 0.14 0.06 0.05
0.2 0.93 0.03 0.11 0.06 0.05
0.5 0.94 0.05 0.07 0.06 0.03
empty 0.0 0.94 0.02 0.09 0.04 0.04
0.1 0.93 0.03 0.11 0.05 0.05
0.2 0.96 0.04 0.05 0.04 0.03
0.5 0.95 0.04 0.07 0.05 0.02
ham 0.0 0.95 0.02 0.09 0.04 0.04
0.1 0.95 0.02 0.08 0.04 0.04
0.2 0.95 0.02 0.09 0.04 0.03
0.5 0.95 0.02 0.08 0.04 0.02
focussed 0.0 0.95 0.02 0.08 0.04 0.04
0.1 0.93 0.01 0.13 0.05 0.05
0.2 0.94 0.03 0.10 0.05 0.04
0.5 0.95 0.03 0.08 0.05 0.02
naive bayes dictionary 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.62 0.65 0.11 0.47 0.42
0.2 0.59 0.73 0.10 0.52 0.42
0.5 0.52 0.87 0.09 0.61 0.31
empty 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.89 0.04 0.19 0.09 0.08
0.2 0.89 0.05 0.16 0.09 0.07
0.5 0.81 0.21 0.17 0.20 0.10
ham 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.04 0.22 0.10 0.09
0.2 0.87 0.05 0.20 0.10 0.08
0.5 0.78 0.24 0.20 0.23 0.11
focussed 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.03 0.23 0.10 0.09
0.2 0.86 0.08 0.19 0.12 0.09
0.5 0.42 1.00 0.17 0.72 0.36
15 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.66 0.83
empty 0.0 0.96 0.00 0.07 0.02 0.03
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.10 0.04 0.03
0.5 0.91 0.00 0.19 0.06 0.03
ham 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.96 0.00 0.08 0.03 0.03
0.2 0.95 0.00 0.09 0.03 0.03
0.5 0.50 1.00 0.00 0.67 0.83
focussed 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.96 0.00 0.08 0.03 0.02
0.2 0.96 0.00 0.07 0.03 0.02
0.5 0.50 1.00 0.00 0.67 0.83
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
empty 0.0 0.99 0.01 0.02 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
ham 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.98 0.02 0.02 0.02 0.00
focussed 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.00 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.54 0.92 0.01 0.62 0.55
0.2 0.50 1.00 0.00 0.66 0.53
0.5 0.50 1.00 0.00 0.67 0.33
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.96 0.05 0.04 0.05 0.04
0.2 0.95 0.05 0.05 0.05 0.04
0.5 0.92 0.10 0.06 0.09 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.06 0.06 0.05
0.2 0.93 0.09 0.06 0.08 0.06
0.5 0.78 0.43 0.02 0.29 0.14
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.07 0.03 0.05 0.04
0.2 0.94 0.07 0.04 0.06 0.05
0.5 0.90 0.17 0.02 0.12 0.06
trec2007-1607252257 adaline dictionary 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.66 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.10 0.04 0.04
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.89 0.01 0.22 0.08 0.04
ham 0.0 0.95 0.01 0.08 0.03 0.03
0.1 0.91 0.01 0.16 0.06 0.05
0.2 0.90 0.01 0.19 0.07 0.05
0.5 0.88 0.01 0.22 0.08 0.04
focussed 0.0 0.95 0.01 0.09 0.03 0.03
0.1 0.95 0.01 0.09 0.04 0.03
0.2 0.95 0.01 0.09 0.04 0.03
0.5 0.95 0.01 0.09 0.04 0.02
logistic regression dictionary 0.0 0.98 0.01 0.02 0.01 0.01
0.1 0.97 0.06 0.01 0.04 0.03
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.02 0.02 0.02 0.00
empty 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.02 0.01 0.01
0.5 0.98 0.01 0.04 0.02 0.00
ham 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.97 0.01 0.05 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
focussed 0.0 0.98 0.02 0.02 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.02 0.01 0.01
0.5 0.98 0.01 0.02 0.02 0.00
naive bayes dictionary 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.54 0.86 0.07 0.60 0.53
0.2 0.52 0.93 0.03 0.63 0.50
0.5 0.53 0.94 0.01 0.63 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.09 0.05 0.08 0.07
0.2 0.91 0.11 0.06 0.09 0.07
0.5 0.87 0.18 0.08 0.15 0.08
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.09 0.08
0.2 0.90 0.12 0.08 0.11 0.08
0.5 0.84 0.22 0.09 0.18 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.10 0.04 0.08 0.07
0.2 0.69 0.62 0.01 0.41 0.33
0.5 0.85 0.24 0.05 0.18 0.09
trec2007-1607252259 adaline dictionary 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.88 0.01 0.23 0.09 0.08
0.2 0.87 0.01 0.25 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.91 0.01 0.17 0.07 0.06
0.2 0.88 0.01 0.22 0.08 0.07
0.5 0.79 0.01 0.41 0.14 0.07
ham 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.89 0.02 0.21 0.08 0.07
0.2 0.84 0.02 0.30 0.11 0.09
0.5 0.76 0.02 0.47 0.17 0.08
focussed 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.90 0.02 0.19 0.07 0.07
0.2 0.84 0.02 0.29 0.11 0.09
0.5 0.82 0.02 0.34 0.12 0.06
logistic regression dictionary 0.0 0.96 0.03 0.06 0.04 0.04
0.1 0.94 0.05 0.07 0.06 0.05
0.2 0.92 0.13 0.03 0.09 0.27
0.5 0.95 0.05 0.06 0.05 0.03
empty 0.0 0.95 0.02 0.07 0.04 0.04
0.1 0.95 0.05 0.05 0.05 0.04
0.2 0.90 0.01 0.20 0.07 0.06
0.5 0.95 0.02 0.08 0.04 0.02
ham 0.0 0.95 0.03 0.07 0.04 0.04
0.1 0.92 0.01 0.15 0.06 0.05
0.2 0.95 0.02 0.09 0.04 0.03
0.5 0.95 0.05 0.05 0.05 0.02
focussed 0.0 0.95 0.03 0.07 0.04 0.05
0.1 0.95 0.03 0.08 0.05 0.04
0.2 0.95 0.05 0.06 0.05 0.04
0.5 0.94 0.03 0.08 0.05 0.02
naive bayes dictionary 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.62 0.64 0.11 0.46 0.42
0.2 0.58 0.73 0.10 0.52 0.42
0.5 0.52 0.88 0.09 0.61 0.31
empty 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.89 0.04 0.18 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.81 0.23 0.15 0.20 0.10
ham 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.04 0.22 0.10 0.09
0.2 0.87 0.05 0.21 0.11 0.08
0.5 0.79 0.23 0.20 0.22 0.11
focussed 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.04 0.21 0.10 0.08
0.2 0.91 0.08 0.11 0.09 0.07
0.5 0.42 1.00 0.17 0.72 0.36
16 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.50 1.00 0.00 0.66 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.96 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.11 0.04 0.03
0.5 0.90 0.00 0.19 0.07 0.03
ham 0.0 0.97 0.00 0.07 0.02 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.09 0.03 0.03
0.5 0.50 1.00 0.00 0.67 0.83
focussed 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.96 0.00 0.08 0.03 0.02
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.02 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
empty 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
ham 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.02 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.01 0.01 0.00
focussed 0.0 0.99 0.01 0.02 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.00 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.54 0.92 0.01 0.61 0.55
0.2 0.50 1.00 0.00 0.66 0.53
0.5 0.50 1.00 0.00 0.66 0.34
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.96 0.05 0.04 0.05 0.04
0.2 0.95 0.06 0.05 0.05 0.04
0.5 0.92 0.10 0.07 0.09 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.05 0.06 0.05
0.2 0.93 0.09 0.05 0.08 0.06
0.5 0.79 0.41 0.02 0.28 0.14
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.96 0.06 0.03 0.05 0.04
0.2 0.94 0.07 0.04 0.06 0.05
0.5 0.86 0.25 0.03 0.17 0.09
trec2007-1607252257 adaline dictionary 0.0 0.95 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.67 0.69
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.95 0.01 0.09 0.03 0.03
0.1 0.94 0.01 0.11 0.04 0.04
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.89 0.01 0.22 0.08 0.04
ham 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.91 0.01 0.16 0.06 0.06
0.2 0.91 0.01 0.18 0.07 0.05
0.5 0.88 0.01 0.22 0.08 0.04
focussed 0.0 0.95 0.01 0.09 0.03 0.03
0.1 0.95 0.01 0.09 0.04 0.03
0.2 0.95 0.01 0.09 0.04 0.03
0.5 0.95 0.01 0.09 0.04 0.02
logistic regression dictionary 0.0 0.98 0.01 0.03 0.01 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.01 0.01
0.5 0.98 0.01 0.03 0.02 0.00
empty 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
ham 0.0 0.98 0.02 0.02 0.02 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
focussed 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.02 0.02 0.00
naive bayes dictionary 0.0 0.94 0.10 0.03 0.07 0.07
0.1 0.53 0.88 0.06 0.61 0.54
0.2 0.52 0.93 0.03 0.63 0.50
0.5 0.53 0.93 0.01 0.62 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.10 0.04 0.08 0.07
0.2 0.92 0.10 0.06 0.09 0.07
0.5 0.87 0.18 0.08 0.15 0.07
ham 0.0 0.94 0.08 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.09 0.08
0.2 0.90 0.11 0.08 0.10 0.08
0.5 0.84 0.23 0.10 0.19 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.11 0.06 0.09 0.08
0.2 0.57 0.77 0.08 0.54 0.43
0.5 0.54 0.91 0.00 0.61 0.30
trec2007-1607252259 adaline dictionary 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.87 0.01 0.24 0.09 0.08
0.2 0.88 0.01 0.23 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.01 0.14 0.06 0.06
0.1 0.90 0.01 0.19 0.07 0.06
0.2 0.88 0.01 0.22 0.08 0.06
0.5 0.79 0.01 0.41 0.14 0.07
ham 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.02 0.22 0.08 0.08
0.2 0.84 0.02 0.31 0.11 0.09
0.5 0.76 0.02 0.47 0.17 0.08
focussed 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.90 0.02 0.19 0.08 0.07
0.2 0.88 0.01 0.22 0.08 0.06
0.5 0.85 0.01 0.30 0.10 0.05
logistic regression dictionary 0.0 0.91 0.01 0.17 0.06 0.06
0.1 0.94 0.04 0.07 0.05 0.04
0.2 0.92 0.03 0.13 0.06 0.05
0.5 0.94 0.04 0.08 0.05 0.02
empty 0.0 0.95 0.02 0.08 0.04 0.04
0.1 0.95 0.03 0.06 0.04 0.04
0.2 0.95 0.03 0.08 0.04 0.03
0.5 0.95 0.01 0.09 0.04 0.02
ham 0.0 0.94 0.02 0.10 0.05 0.05
0.1 0.93 0.02 0.11 0.05 0.04
0.2 0.90 0.01 0.20 0.07 0.06
0.5 0.95 0.02 0.08 0.04 0.02
focussed 0.0 0.95 0.02 0.07 0.04 0.04
0.1 0.93 0.02 0.11 0.05 0.05
0.2 0.94 0.05 0.07 0.06 0.05
0.5 0.95 0.03 0.07 0.04 0.02
naive bayes dictionary 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.63 0.64 0.11 0.46 0.42
0.2 0.58 0.73 0.11 0.52 0.42
0.5 0.51 0.88 0.09 0.62 0.31
empty 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.89 0.04 0.17 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.80 0.23 0.17 0.21 0.10
ham 0.0 0.90 0.05 0.16 0.09 0.08
0.1 0.87 0.04 0.22 0.10 0.09
0.2 0.87 0.05 0.21 0.11 0.08
0.5 0.79 0.23 0.19 0.21 0.11
focussed 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.76 0.38 0.10 0.29 0.26
0.2 0.86 0.04 0.25 0.11 0.09
0.5 0.43 0.99 0.15 0.71 0.35
17 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.50 1.00 0.00 0.66 0.70
0.2 0.50 1.00 0.00 0.66 0.73
0.5 0.50 1.00 0.00 0.66 0.83
empty 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.96 0.00 0.09 0.03 0.03
0.2 0.94 0.00 0.11 0.04 0.03
0.5 0.91 0.00 0.19 0.06 0.03
ham 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.96 0.00 0.08 0.03 0.03
0.2 0.95 0.00 0.09 0.03 0.03
0.5 0.50 1.00 0.00 0.66 0.84
focussed 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.96 0.00 0.07 0.03 0.02
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.96 0.00 0.09 0.03 0.01
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.01 0.01 0.00
empty 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.00 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
ham 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.00 0.02 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
focussed 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.53 0.92 0.01 0.62 0.55
0.2 0.50 1.00 0.00 0.66 0.53
0.5 0.50 1.00 0.00 0.67 0.33
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.96 0.05 0.04 0.05 0.04
0.2 0.95 0.06 0.05 0.05 0.04
0.5 0.92 0.10 0.06 0.09 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.05 0.06 0.05
0.2 0.92 0.09 0.06 0.08 0.06
0.5 0.77 0.44 0.01 0.30 0.15
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.06 0.04 0.05 0.04
0.2 0.94 0.10 0.02 0.07 0.06
0.5 0.91 0.13 0.06 0.11 0.05
trec2007-1607252257 adaline dictionary 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.66 0.84
empty 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.10 0.04 0.03
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.89 0.01 0.22 0.08 0.04
ham 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.92 0.01 0.16 0.06 0.05
0.2 0.90 0.01 0.19 0.07 0.05
0.5 0.88 0.01 0.22 0.08 0.04
focussed 0.0 0.95 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.10 0.04 0.03
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
logistic regression dictionary 0.0 0.98 0.01 0.03 0.01 0.01
0.1 0.98 0.01 0.02 0.01 0.01
0.2 0.98 0.01 0.02 0.01 0.01
0.5 0.98 0.01 0.03 0.02 0.00
empty 0.0 0.98 0.01 0.02 0.01 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.02 0.02 0.02 0.01
ham 0.0 0.98 0.01 0.02 0.01 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.02 0.02 0.02 0.01
0.5 0.98 0.01 0.04 0.02 0.00
focussed 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
naive bayes dictionary 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.54 0.86 0.07 0.60 0.53
0.2 0.52 0.93 0.04 0.63 0.50
0.5 0.53 0.93 0.01 0.62 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.09 0.05 0.08 0.07
0.2 0.92 0.11 0.06 0.09 0.07
0.5 0.87 0.19 0.08 0.15 0.08
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.08 0.08
0.2 0.90 0.11 0.09 0.11 0.09
0.5 0.84 0.23 0.10 0.18 0.09
focussed 0.0 0.94 0.08 0.04 0.07 0.07
0.1 0.92 0.10 0.05 0.09 0.08
0.2 0.86 0.26 0.02 0.18 0.14
0.5 0.60 0.75 0.06 0.52 0.26
trec2007-1607252259 adaline dictionary 0.0 0.92 0.02 0.14 0.06 0.06
0.1 0.88 0.01 0.23 0.08 0.08
0.2 0.87 0.01 0.24 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.91 0.01 0.17 0.07 0.06
0.2 0.88 0.01 0.22 0.08 0.06
0.5 0.79 0.01 0.41 0.14 0.07
ham 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.89 0.02 0.21 0.08 0.07
0.2 0.83 0.02 0.32 0.12 0.09
0.5 0.76 0.02 0.46 0.17 0.08
focussed 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.89 0.02 0.21 0.08 0.07
0.2 0.82 0.01 0.36 0.13 0.10
0.5 0.79 0.02 0.41 0.15 0.08
logistic regression dictionary 0.0 0.95 0.02 0.08 0.04 0.04
0.1 0.94 0.06 0.06 0.06 0.05
0.2 0.94 0.03 0.08 0.05 0.04
0.5 0.94 0.03 0.10 0.05 0.02
empty 0.0 0.93 0.01 0.14 0.05 0.05
0.1 0.95 0.01 0.09 0.04 0.03
0.2 0.93 0.01 0.12 0.05 0.04
0.5 0.95 0.02 0.08 0.04 0.02
ham 0.0 0.92 0.03 0.13 0.06 0.06
0.1 0.95 0.02 0.07 0.04 0.04
0.2 0.95 0.03 0.08 0.05 0.03
0.5 0.95 0.04 0.06 0.05 0.02
focussed 0.0 0.95 0.05 0.05 0.05 0.05
0.1 0.94 0.03 0.08 0.05 0.04
0.2 0.95 0.04 0.06 0.05 0.04
0.5 0.95 0.02 0.09 0.04 0.02
naive bayes dictionary 0.0 0.90 0.05 0.16 0.08 0.08
0.1 0.63 0.64 0.11 0.46 0.42
0.2 0.58 0.73 0.10 0.52 0.42
0.5 0.52 0.87 0.10 0.61 0.31
empty 0.0 0.90 0.05 0.16 0.08 0.08
0.1 0.89 0.04 0.17 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.80 0.22 0.17 0.20 0.10
ham 0.0 0.90 0.04 0.17 0.08 0.08
0.1 0.87 0.04 0.21 0.10 0.09
0.2 0.87 0.05 0.20 0.10 0.08
0.5 0.79 0.22 0.20 0.22 0.11
focussed 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.84 0.03 0.30 0.12 0.11
0.2 0.86 0.05 0.24 0.11 0.09
0.5 0.71 0.32 0.25 0.30 0.14
18 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.66 0.84
empty 0.0 0.96 0.00 0.07 0.02 0.03
0.1 0.96 0.00 0.08 0.03 0.03
0.2 0.95 0.00 0.11 0.04 0.03
0.5 0.90 0.00 0.19 0.07 0.03
ham 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.10 0.03 0.03
0.5 0.50 1.00 0.00 0.67 0.83
focussed 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.96 0.00 0.07 0.02 0.02
0.2 0.96 0.00 0.07 0.03 0.02
0.5 0.50 1.00 0.00 0.66 0.83
logistic regression dictionary 0.0 0.99 0.00 0.02 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
empty 0.0 0.99 0.01 0.02 0.01 0.00
0.1 0.99 0.01 0.02 0.01 0.00
0.2 0.99 0.00 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
ham 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.02 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
focussed 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.53 0.92 0.01 0.62 0.55
0.2 0.50 1.00 0.00 0.66 0.53
0.5 0.50 1.00 0.00 0.66 0.33
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.96 0.05 0.04 0.04 0.04
0.2 0.94 0.06 0.05 0.06 0.04
0.5 0.92 0.10 0.06 0.09 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.06 0.06 0.05
0.2 0.93 0.09 0.06 0.08 0.06
0.5 0.77 0.44 0.02 0.30 0.15
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.82 0.36 0.00 0.24 0.22
0.2 0.94 0.08 0.05 0.07 0.05
0.5 0.89 0.21 0.01 0.14 0.07
trec2007-1607252257 adaline dictionary 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.66 0.74
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.10 0.04 0.04
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.89 0.01 0.21 0.08 0.04
ham 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.91 0.01 0.17 0.06 0.05
0.2 0.90 0.01 0.18 0.07 0.05
0.5 0.88 0.01 0.22 0.08 0.04
focussed 0.0 0.95 0.01 0.08 0.03 0.03
0.1 0.94 0.01 0.12 0.04 0.04
0.2 0.94 0.01 0.11 0.04 0.03
0.5 0.50 1.00 0.00 0.66 0.83
logistic regression dictionary 0.0 0.98 0.02 0.03 0.02 0.01
0.1 0.98 0.01 0.03 0.01 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.02 0.02 0.02 0.00
empty 0.0 0.98 0.01 0.02 0.01 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
ham 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
focussed 0.0 0.98 0.01 0.02 0.01 0.01
0.1 0.98 0.01 0.03 0.01 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.01 0.04 0.02 0.00
naive bayes dictionary 0.0 0.94 0.09 0.04 0.07 0.07
0.1 0.52 0.88 0.07 0.61 0.54
0.2 0.52 0.93 0.03 0.63 0.51
0.5 0.53 0.94 0.01 0.63 0.31
empty 0.0 0.94 0.10 0.03 0.07 0.07
0.1 0.93 0.10 0.04 0.08 0.07
0.2 0.91 0.11 0.06 0.10 0.07
0.5 0.87 0.19 0.08 0.15 0.07
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.09 0.08
0.2 0.90 0.12 0.08 0.10 0.08
0.5 0.84 0.22 0.09 0.18 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.11 0.05 0.09 0.08
0.2 0.73 0.53 0.00 0.35 0.29
0.5 0.81 0.34 0.03 0.24 0.12
trec2007-1607252259 adaline dictionary 0.0 0.92 0.02 0.14 0.06 0.06
0.1 0.88 0.01 0.24 0.09 0.08
0.2 0.87 0.01 0.25 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.91 0.01 0.17 0.07 0.06
0.2 0.89 0.01 0.21 0.08 0.06
0.5 0.79 0.01 0.40 0.14 0.07
ham 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.02 0.22 0.08 0.07
0.2 0.84 0.02 0.31 0.11 0.09
0.5 0.76 0.02 0.46 0.17 0.08
focussed 0.0 0.92 0.01 0.14 0.06 0.06
0.1 0.87 0.00 0.26 0.09 0.08
0.2 0.85 0.01 0.30 0.11 0.08
0.5 0.81 0.01 0.37 0.13 0.06
logistic regression dictionary 0.0 0.94 0.09 0.04 0.07 0.07
0.1 0.94 0.03 0.10 0.05 0.05
0.2 0.94 0.04 0.09 0.05 0.04
0.5 0.94 0.04 0.09 0.05 0.03
empty 0.0 0.95 0.05 0.05 0.05 0.05
0.1 0.95 0.04 0.06 0.05 0.04
0.2 0.96 0.03 0.05 0.04 0.03
0.5 0.95 0.04 0.05 0.04 0.02
ham 0.0 0.96 0.03 0.06 0.04 0.04
0.1 0.95 0.07 0.04 0.06 0.05
0.2 0.95 0.03 0.07 0.04 0.03
0.5 0.93 0.02 0.12 0.05 0.02
focussed 0.0 0.95 0.02 0.08 0.04 0.04
0.1 0.95 0.03 0.07 0.04 0.04
0.2 0.95 0.05 0.06 0.05 0.04
0.5 0.95 0.03 0.06 0.04 0.02
naive bayes dictionary 0.0 0.90 0.05 0.15 0.08 0.08
0.1 0.63 0.64 0.10 0.46 0.42
0.2 0.59 0.73 0.10 0.52 0.42
0.5 0.52 0.87 0.09 0.61 0.31
empty 0.0 0.90 0.04 0.17 0.08 0.08
0.1 0.89 0.04 0.17 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.80 0.22 0.17 0.21 0.10
ham 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.04 0.22 0.10 0.09
0.2 0.87 0.05 0.20 0.10 0.08
0.5 0.78 0.24 0.20 0.22 0.11
focussed 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.89 0.04 0.17 0.08 0.17
0.2 0.86 0.05 0.24 0.11 0.09
0.5 0.78 0.32 0.12 0.25 0.13
19 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.66 0.83
empty 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.96 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.11 0.04 0.03
0.5 0.90 0.00 0.19 0.07 0.03
ham 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.10 0.04 0.03
0.5 0.50 1.00 0.00 0.67 0.83
focussed 0.0 0.96 0.00 0.07 0.02 0.03
0.1 0.96 0.00 0.07 0.03 0.02
0.2 0.96 0.00 0.07 0.03 0.02
0.5 0.96 0.00 0.07 0.03 0.01
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
empty 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.00 0.01 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
ham 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
focussed 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.53 0.92 0.01 0.62 0.55
0.2 0.50 1.00 0.00 0.67 0.53
0.5 0.50 1.00 0.00 0.67 0.33
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.96 0.05 0.04 0.05 0.04
0.2 0.95 0.06 0.05 0.06 0.04
0.5 0.92 0.10 0.06 0.09 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.05 0.06 0.05
0.2 0.93 0.09 0.06 0.08 0.06
0.5 0.78 0.42 0.02 0.29 0.14
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.06 0.03 0.05 0.04
0.2 0.95 0.07 0.04 0.06 0.05
0.5 0.90 0.17 0.03 0.12 0.06
trec2007-1607252257 adaline dictionary 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.95 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.10 0.04 0.04
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.89 0.01 0.22 0.08 0.04
ham 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.91 0.01 0.17 0.06 0.05
0.2 0.90 0.01 0.19 0.07 0.05
0.5 0.88 0.01 0.22 0.08 0.04
focussed 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.09 0.04 0.03
0.2 0.95 0.01 0.09 0.03 0.03
0.5 0.94 0.01 0.10 0.04 0.02
logistic regression dictionary 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.01 0.04 0.02 0.01
0.2 0.98 0.02 0.02 0.02 0.01
0.5 0.98 0.01 0.04 0.02 0.00
empty 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.02 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.01 0.01
0.5 0.98 0.01 0.03 0.02 0.00
ham 0.0 0.98 0.01 0.02 0.01 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
focussed 0.0 0.98 0.01 0.02 0.02 0.01
0.1 0.98 0.01 0.04 0.02 0.01
0.2 0.98 0.02 0.02 0.02 0.01
0.5 0.98 0.02 0.02 0.02 0.00
naive bayes dictionary 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.54 0.86 0.07 0.59 0.53
0.2 0.52 0.93 0.03 0.63 0.50
0.5 0.53 0.93 0.01 0.62 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.10 0.05 0.08 0.07
0.2 0.92 0.11 0.06 0.09 0.07
0.5 0.87 0.18 0.08 0.15 0.08
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.05 0.08 0.08
0.2 0.90 0.12 0.08 0.11 0.08
0.5 0.85 0.22 0.09 0.17 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.11 0.04 0.09 0.08
0.2 0.91 0.13 0.06 0.10 0.08
0.5 0.61 0.76 0.01 0.51 0.26
trec2007-1607252259 adaline dictionary 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.01 0.24 0.09 0.08
0.2 0.87 0.01 0.25 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.91 0.01 0.17 0.07 0.06
0.2 0.89 0.01 0.21 0.08 0.06
0.5 0.79 0.01 0.41 0.14 0.07
ham 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.88 0.01 0.22 0.08 0.08
0.2 0.84 0.02 0.31 0.12 0.09
0.5 0.76 0.02 0.47 0.17 0.08
focussed 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.86 0.02 0.27 0.10 0.09
0.2 0.86 0.01 0.27 0.10 0.08
0.5 0.87 0.02 0.24 0.09 0.05
logistic regression dictionary 0.0 0.95 0.05 0.04 0.05 0.05
0.1 0.94 0.04 0.09 0.06 0.05
0.2 0.92 0.02 0.14 0.06 0.05
0.5 0.94 0.03 0.08 0.05 0.02
empty 0.0 0.95 0.05 0.05 0.05 0.05
0.1 0.95 0.04 0.05 0.04 0.04
0.2 0.95 0.03 0.07 0.04 0.03
0.5 0.94 0.02 0.11 0.05 0.02
ham 0.0 0.96 0.04 0.05 0.04 0.04
0.1 0.95 0.02 0.07 0.04 0.04
0.2 0.94 0.02 0.10 0.05 0.03
0.5 0.95 0.04 0.06 0.04 0.02
focussed 0.0 0.95 0.04 0.06 0.04 0.04
0.1 0.94 0.04 0.08 0.05 0.05
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.95 0.02 0.07 0.04 0.02
naive bayes dictionary 0.0 0.90 0.05 0.16 0.08 0.08
0.1 0.62 0.64 0.11 0.46 0.42
0.2 0.58 0.73 0.11 0.52 0.41
0.5 0.52 0.87 0.10 0.61 0.31
empty 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.89 0.04 0.18 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.80 0.23 0.17 0.21 0.10
ham 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.87 0.04 0.21 0.10 0.09
0.2 0.87 0.05 0.21 0.10 0.09
0.5 0.79 0.23 0.19 0.21 0.11
focussed 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.85 0.04 0.25 0.11 0.10
0.2 0.92 0.04 0.11 0.07 0.05
0.5 0.53 0.85 0.09 0.60 0.30
20 trec2007-1607201347 adaline dictionary 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.50 1.00 0.00 0.66 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.66 0.83
empty 0.0 0.96 0.00 0.07 0.03 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.10 0.04 0.03
0.5 0.90 0.00 0.19 0.07 0.03
ham 0.0 0.96 0.00 0.07 0.02 0.02
0.1 0.95 0.00 0.09 0.03 0.03
0.2 0.95 0.00 0.09 0.03 0.03
0.5 0.50 1.00 0.00 0.67 0.83
focussed 0.0 0.97 0.00 0.06 0.02 0.03
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.95 0.00 0.09 0.03 0.03
0.5 0.50 1.00 0.00 0.67 0.83
logistic regression dictionary 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
empty 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.00 0.01 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
ham 0.0 0.99 0.00 0.02 0.01 0.00
0.1 0.99 0.01 0.02 0.01 0.00
0.2 0.99 0.01 0.01 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
focussed 0.0 0.99 0.01 0.01 0.01 0.00
0.1 0.99 0.01 0.01 0.01 0.00
0.2 0.99 0.01 0.02 0.01 0.00
0.5 0.99 0.01 0.02 0.01 0.00
naive bayes dictionary 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.54 0.92 0.01 0.62 0.55
0.2 0.50 1.00 0.00 0.66 0.53
0.5 0.50 1.00 0.00 0.67 0.33
empty 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.95 0.05 0.04 0.05 0.04
0.2 0.95 0.06 0.05 0.06 0.05
0.5 0.92 0.10 0.06 0.09 0.04
ham 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.94 0.06 0.06 0.06 0.05
0.2 0.93 0.09 0.06 0.08 0.06
0.5 0.78 0.43 0.02 0.29 0.14
focussed 0.0 0.96 0.05 0.03 0.04 0.04
0.1 0.96 0.06 0.03 0.05 0.05
0.2 0.93 0.08 0.05 0.07 0.06
0.5 0.91 0.14 0.03 0.10 0.05
trec2007-1607252257 adaline dictionary 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.50 1.00 0.00 0.67 0.70
0.2 0.50 1.00 0.00 0.67 0.73
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.94 0.01 0.10 0.04 0.04
0.2 0.93 0.01 0.13 0.05 0.04
0.5 0.89 0.01 0.21 0.08 0.04
ham 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.91 0.01 0.17 0.06 0.05
0.2 0.90 0.01 0.18 0.07 0.05
0.5 0.88 0.01 0.22 0.08 0.04
focussed 0.0 0.96 0.01 0.08 0.03 0.03
0.1 0.95 0.01 0.09 0.03 0.03
0.2 0.94 0.01 0.11 0.04 0.03
0.5 0.50 1.00 0.00 0.67 0.83
logistic regression dictionary 0.0 0.98 0.01 0.04 0.02 0.01
0.1 0.98 0.01 0.02 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
empty 0.0 0.98 0.02 0.02 0.02 0.01
0.1 0.98 0.01 0.02 0.01 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
ham 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.01 0.03 0.01 0.01
0.2 0.98 0.01 0.02 0.02 0.01
0.5 0.98 0.01 0.03 0.02 0.00
focussed 0.0 0.98 0.01 0.03 0.02 0.01
0.1 0.98 0.01 0.03 0.02 0.01
0.2 0.98 0.01 0.03 0.02 0.01
0.5 0.98 0.03 0.02 0.03 0.01
naive bayes dictionary 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.53 0.86 0.07 0.60 0.54
0.2 0.52 0.93 0.03 0.63 0.51
0.5 0.53 0.94 0.01 0.62 0.31
empty 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.10 0.04 0.08 0.07
0.2 0.91 0.11 0.06 0.09 0.07
0.5 0.87 0.19 0.08 0.15 0.08
ham 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.92 0.10 0.06 0.09 0.08
0.2 0.90 0.11 0.08 0.10 0.08
0.5 0.84 0.23 0.09 0.18 0.09
focussed 0.0 0.94 0.09 0.03 0.07 0.07
0.1 0.93 0.11 0.04 0.09 0.08
0.2 0.91 0.14 0.04 0.11 0.09
0.5 0.61 0.77 0.02 0.52 0.26
trec2007-1607252259 adaline dictionary 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.01 0.23 0.09 0.08
0.2 0.87 0.01 0.24 0.09 0.07
0.5 0.50 1.00 0.00 0.67 0.83
empty 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.91 0.01 0.17 0.07 0.06
0.2 0.89 0.01 0.21 0.08 0.06
0.5 0.79 0.01 0.41 0.14 0.07
ham 0.0 0.92 0.01 0.15 0.06 0.06
0.1 0.88 0.02 0.22 0.08 0.07
0.2 0.84 0.02 0.30 0.11 0.09
0.5 0.76 0.02 0.46 0.17 0.08
focussed 0.0 0.92 0.02 0.15 0.06 0.06
0.1 0.87 0.02 0.23 0.09 0.08
0.2 0.87 0.01 0.24 0.09 0.07
0.5 0.77 0.02 0.45 0.16 0.08
logistic regression dictionary 0.0 0.94 0.02 0.10 0.05 0.05
0.1 0.94 0.05 0.06 0.06 0.05
0.2 0.93 0.03 0.10 0.06 0.04
0.5 0.92 0.02 0.13 0.06 0.03
empty 0.0 0.95 0.04 0.05 0.05 0.04
0.1 0.95 0.02 0.08 0.04 0.04
0.2 0.95 0.03 0.07 0.04 0.03
0.5 0.95 0.02 0.08 0.04 0.02
ham 0.0 0.95 0.02 0.08 0.04 0.04
0.1 0.95 0.04 0.06 0.04 0.04
0.2 0.94 0.03 0.09 0.05 0.04
0.5 0.95 0.04 0.06 0.04 0.02
focussed 0.0 0.95 0.03 0.06 0.04 0.04
0.1 0.95 0.02 0.08 0.04 0.04
0.2 0.95 0.02 0.09 0.04 0.03
0.5 0.94 0.03 0.10 0.05 0.03
naive bayes dictionary 0.0 0.90 0.05 0.16 0.08 0.08
0.1 0.63 0.64 0.11 0.46 0.42
0.2 0.58 0.72 0.11 0.52 0.42
0.5 0.51 0.87 0.10 0.61 0.31
empty 0.0 0.90 0.04 0.16 0.08 0.08
0.1 0.89 0.04 0.18 0.09 0.08
0.2 0.89 0.05 0.17 0.09 0.07
0.5 0.81 0.22 0.17 0.20 0.10
ham 0.0 0.90 0.05 0.16 0.08 0.08
0.1 0.87 0.04 0.22 0.10 0.09
0.2 0.87 0.05 0.20 0.10 0.08
0.5 0.79 0.22 0.20 0.21 0.11
focussed 0.0 0.90 0.04 0.17 0.08 0.08
0.1 0.89 0.04 0.19 0.09 0.08
0.2 0.60 0.70 0.10 0.50 0.40
0.5 0.54 0.73 0.19 0.55 0.28
Display the source blob
Display the rendered blob
Raw
1608030254.ipynb
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<style>div.container { width:100% !important; }</style>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"<style>div.cell { width:65% !important; margin: auto; }</style>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"<style>div.output_png { background-color: white; position: relative; width: 400% !important; left: -35%; overflow-x: visible !important; }</style>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"<style>div.output_png img { max-width: 1900px !important; margin: auto !important; }</style>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"import numpy as np\n",
"import pandas as pd\n",
"import matplotlib.pyplot as plt\n",
"%matplotlib inline\n",
"\n",
"from IPython.core.display import display, HTML\n",
"display(HTML(\"<style>div.container { width:100% !important; }</style>\"))\n",
"display(HTML(\"<style>div.cell { width:65% !important; margin: auto; }</style>\"))\n",
"display(HTML(\"<style>div.output_png { background-color: white; position: relative; width: 400% !important; left: -35%; overflow-x: visible !important; }</style>\"))\n",
"display(HTML(\"<style>div.output_png img { max-width: 1900px !important; margin: auto !important; }</style>\"))\n",
"\n",
"%load_ext autoreload\n",
"%autoreload 1\n",
"\n",
"import sys\n",
"import os.path\n",
"absolute_path_code_dir = os.path.join('/', 'home', 'justine', 'Dropbox', 'imperial', 'computing', 'thesis', 'code')\n",
"sys.path.insert(0, absolute_path_code_dir)\n",
"\n",
"from helpers.i_o import join_repetitions\n",
"from analyse.visualise_offline import visualise"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"collapsed": false,
"scrolled": true
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th></th>\n",
" <th>metrics</th>\n",
" <th>AUC</th>\n",
" <th>FNR</th>\n",
" <th>FPR</th>\n",
" <th>error_test</th>\n",
" <th>error_train</th>\n",
" </tr>\n",
" <tr>\n",
" <th>classifier</th>\n",
" <th>attack</th>\n",
" <th>% poisoned</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th rowspan=\"12\" valign=\"top\">adaline</th>\n",
" <th rowspan=\"4\" valign=\"top\">dictionary</th>\n",
" <th>0.0</th>\n",
" <td>0.963254</td>\n",
" <td>0.002713</td>\n",
" <td>0.070779</td>\n",
" <td>0.025508</td>\n",
" <td>0.024275</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.1</th>\n",
" <td>0.500000</td>\n",
" <td>1.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.665426</td>\n",
" <td>0.699100</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.2</th>\n",
" <td>0.500000</td>\n",
" <td>1.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.665822</td>\n",
" <td>0.732133</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.5</th>\n",
" <td>0.500000</td>\n",
" <td>1.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.665389</td>\n",
" <td>0.832944</td>\n",
" </tr>\n",
" <tr>\n",
" <th rowspan=\"4\" valign=\"top\">empty</th>\n",
" <th>0.0</th>\n",
" <td>0.963843</td>\n",
" <td>0.002669</td>\n",
" <td>0.069645</td>\n",
" <td>0.025032</td>\n",
" <td>0.024499</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.1</th>\n",
" <td>0.955359</td>\n",
" <td>0.002324</td>\n",
" <td>0.086958</td>\n",
" <td>0.030603</td>\n",
" <td>0.026982</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.2</th>\n",
" <td>0.945653</td>\n",
" <td>0.002416</td>\n",
" <td>0.106277</td>\n",
" <td>0.037199</td>\n",
" <td>0.028995</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.5</th>\n",
" <td>0.903387</td>\n",
" <td>0.003322</td>\n",
" <td>0.189904</td>\n",
" <td>0.065754</td>\n",
" <td>0.032263</td>\n",
" </tr>\n",
" <tr>\n",
" <th rowspan=\"4\" valign=\"top\">ham</th>\n",
" <th>0.0</th>\n",
" <td>0.963970</td>\n",
" <td>0.002735</td>\n",
" <td>0.069326</td>\n",
" <td>0.024999</td>\n",
" <td>0.024446</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.1</th>\n",
" <td>0.953514</td>\n",
" <td>0.003805</td>\n",
" <td>0.089167</td>\n",
" <td>0.032386</td>\n",
" <td>0.028864</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.2</th>\n",
" <td>0.950158</td>\n",
" <td>0.004105</td>\n",
" <td>0.095578</td>\n",
" <td>0.034675</td>\n",
" <td>0.027392</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.5</th>\n",
" <td>0.500000</td>\n",
" <td>1.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.665813</td>\n",
" <td>0.832644</td>\n",
" </tr>\n",
" <tr>\n",
" <th rowspan=\"12\" valign=\"top\">logistic regression</th>\n",
" <th rowspan=\"4\" valign=\"top\">dictionary</th>\n",
" <th>0.0</th>\n",
" <td>0.990174</td>\n",
" <td>0.006080</td>\n",
" <td>0.013571</td>\n",
" <td>0.008588</td>\n",
" <td>0.001050</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.1</th>\n",
" <td>0.990098</td>\n",
" <td>0.006090</td>\n",
" <td>0.013713</td>\n",
" <td>0.008640</td>\n",
" <td>0.000986</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.2</th>\n",
" <td>0.989215</td>\n",
" <td>0.006250</td>\n",
" <td>0.015319</td>\n",
" <td>0.009283</td>\n",
" <td>0.000895</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.5</th>\n",
" <td>0.987602</td>\n",
" <td>0.007773</td>\n",
" <td>0.017022</td>\n",
" <td>0.010873</td>\n",
" <td>0.000619</td>\n",
" </tr>\n",
" <tr>\n",
" <th rowspan=\"4\" valign=\"top\">empty</th>\n",
" <th>0.0</th>\n",
" <td>0.990052</td>\n",
" <td>0.005894</td>\n",
" <td>0.014001</td>\n",
" <td>0.008599</td>\n",
" <td>0.001189</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.1</th>\n",
" <td>0.990034</td>\n",
" <td>0.006043</td>\n",
" <td>0.013889</td>\n",
" <td>0.008669</td>\n",
" <td>0.001123</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.2</th>\n",
" <td>0.989311</td>\n",
" <td>0.006558</td>\n",
" <td>0.014819</td>\n",
" <td>0.009323</td>\n",
" <td>0.000902</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.5</th>\n",
" <td>0.988231</td>\n",
" <td>0.006550</td>\n",
" <td>0.016988</td>\n",
" <td>0.010044</td>\n",
" <td>0.000585</td>\n",
" </tr>\n",
" <tr>\n",
" <th rowspan=\"4\" valign=\"top\">ham</th>\n",
" <th>0.0</th>\n",
" <td>0.990159</td>\n",
" <td>0.005961</td>\n",
" <td>0.013720</td>\n",
" <td>0.008551</td>\n",
" <td>0.001204</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.1</th>\n",
" <td>0.989255</td>\n",
" <td>0.006521</td>\n",
" <td>0.014969</td>\n",
" <td>0.009345</td>\n",
" <td>0.000935</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.2</th>\n",
" <td>0.988522</td>\n",
" <td>0.007365</td>\n",
" <td>0.015592</td>\n",
" <td>0.010121</td>\n",
" <td>0.001009</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.5</th>\n",
" <td>0.986473</td>\n",
" <td>0.009212</td>\n",
" <td>0.017843</td>\n",
" <td>0.012090</td>\n",
" <td>0.000663</td>\n",
" </tr>\n",
" <tr>\n",
" <th rowspan=\"12\" valign=\"top\">naive bayes</th>\n",
" <th rowspan=\"4\" valign=\"top\">dictionary</th>\n",
" <th>0.0</th>\n",
" <td>0.961845</td>\n",
" <td>0.049294</td>\n",
" <td>0.027015</td>\n",
" <td>0.041820</td>\n",
" <td>0.040084</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.1</th>\n",
" <td>0.534530</td>\n",
" <td>0.920419</td>\n",
" <td>0.010521</td>\n",
" <td>0.616045</td>\n",
" <td>0.553860</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.2</th>\n",
" <td>0.501303</td>\n",
" <td>0.995645</td>\n",
" <td>0.001748</td>\n",
" <td>0.663899</td>\n",
" <td>0.530552</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.5</th>\n",
" <td>0.500000</td>\n",
" <td>1.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.666451</td>\n",
" <td>0.332579</td>\n",
" </tr>\n",
" <tr>\n",
" <th rowspan=\"4\" valign=\"top\">empty</th>\n",
" <th>0.0</th>\n",
" <td>0.962174</td>\n",
" <td>0.048666</td>\n",
" <td>0.026986</td>\n",
" <td>0.041417</td>\n",
" <td>0.040359</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.1</th>\n",
" <td>0.954476</td>\n",
" <td>0.050252</td>\n",
" <td>0.040796</td>\n",
" <td>0.047094</td>\n",
" <td>0.042053</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.2</th>\n",
" <td>0.945844</td>\n",
" <td>0.057143</td>\n",
" <td>0.051168</td>\n",
" <td>0.055149</td>\n",
" <td>0.043369</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.5</th>\n",
" <td>0.918902</td>\n",
" <td>0.098656</td>\n",
" <td>0.063539</td>\n",
" <td>0.086910</td>\n",
" <td>0.042562</td>\n",
" </tr>\n",
" <tr>\n",
" <th rowspan=\"4\" valign=\"top\">ham</th>\n",
" <th>0.0</th>\n",
" <td>0.962132</td>\n",
" <td>0.049105</td>\n",
" <td>0.026631</td>\n",
" <td>0.041587</td>\n",
" <td>0.040464</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.1</th>\n",
" <td>0.941962</td>\n",
" <td>0.059509</td>\n",
" <td>0.056568</td>\n",
" <td>0.058526</td>\n",
" <td>0.051663</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.2</th>\n",
" <td>0.926077</td>\n",
" <td>0.089938</td>\n",
" <td>0.057907</td>\n",
" <td>0.079211</td>\n",
" <td>0.061562</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.5</th>\n",
" <td>0.776466</td>\n",
" <td>0.431486</td>\n",
" <td>0.015581</td>\n",
" <td>0.292404</td>\n",
" <td>0.144814</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
"metrics AUC FNR FPR \\\n",
"classifier attack % poisoned \n",
"adaline dictionary 0.0 0.963254 0.002713 0.070779 \n",
" 0.1 0.500000 1.000000 0.000000 \n",
" 0.2 0.500000 1.000000 0.000000 \n",
" 0.5 0.500000 1.000000 0.000000 \n",
" empty 0.0 0.963843 0.002669 0.069645 \n",
" 0.1 0.955359 0.002324 0.086958 \n",
" 0.2 0.945653 0.002416 0.106277 \n",
" 0.5 0.903387 0.003322 0.189904 \n",
" ham 0.0 0.963970 0.002735 0.069326 \n",
" 0.1 0.953514 0.003805 0.089167 \n",
" 0.2 0.950158 0.004105 0.095578 \n",
" 0.5 0.500000 1.000000 0.000000 \n",
"logistic regression dictionary 0.0 0.990174 0.006080 0.013571 \n",
" 0.1 0.990098 0.006090 0.013713 \n",
" 0.2 0.989215 0.006250 0.015319 \n",
" 0.5 0.987602 0.007773 0.017022 \n",
" empty 0.0 0.990052 0.005894 0.014001 \n",
" 0.1 0.990034 0.006043 0.013889 \n",
" 0.2 0.989311 0.006558 0.014819 \n",
" 0.5 0.988231 0.006550 0.016988 \n",
" ham 0.0 0.990159 0.005961 0.013720 \n",
" 0.1 0.989255 0.006521 0.014969 \n",
" 0.2 0.988522 0.007365 0.015592 \n",
" 0.5 0.986473 0.009212 0.017843 \n",
"naive bayes dictionary 0.0 0.961845 0.049294 0.027015 \n",
" 0.1 0.534530 0.920419 0.010521 \n",
" 0.2 0.501303 0.995645 0.001748 \n",
" 0.5 0.500000 1.000000 0.000000 \n",
" empty 0.0 0.962174 0.048666 0.026986 \n",
" 0.1 0.954476 0.050252 0.040796 \n",
" 0.2 0.945844 0.057143 0.051168 \n",
" 0.5 0.918902 0.098656 0.063539 \n",
" ham 0.0 0.962132 0.049105 0.026631 \n",
" 0.1 0.941962 0.059509 0.056568 \n",
" 0.2 0.926077 0.089938 0.057907 \n",
" 0.5 0.776466 0.431486 0.015581 \n",
"\n",
"metrics error_test error_train \n",
"classifier attack % poisoned \n",
"adaline dictionary 0.0 0.025508 0.024275 \n",
" 0.1 0.665426 0.699100 \n",
" 0.2 0.665822 0.732133 \n",
" 0.5 0.665389 0.832944 \n",
" empty 0.0 0.025032 0.024499 \n",
" 0.1 0.030603 0.026982 \n",
" 0.2 0.037199 0.028995 \n",
" 0.5 0.065754 0.032263 \n",
" ham 0.0 0.024999 0.024446 \n",
" 0.1 0.032386 0.028864 \n",
" 0.2 0.034675 0.027392 \n",
" 0.5 0.665813 0.832644 \n",
"logistic regression dictionary 0.0 0.008588 0.001050 \n",
" 0.1 0.008640 0.000986 \n",
" 0.2 0.009283 0.000895 \n",
" 0.5 0.010873 0.000619 \n",
" empty 0.0 0.008599 0.001189 \n",
" 0.1 0.008669 0.001123 \n",
" 0.2 0.009323 0.000902 \n",
" 0.5 0.010044 0.000585 \n",
" ham 0.0 0.008551 0.001204 \n",
" 0.1 0.009345 0.000935 \n",
" 0.2 0.010121 0.001009 \n",
" 0.5 0.012090 0.000663 \n",
"naive bayes dictionary 0.0 0.041820 0.040084 \n",
" 0.1 0.616045 0.553860 \n",
" 0.2 0.663899 0.530552 \n",
" 0.5 0.666451 0.332579 \n",
" empty 0.0 0.041417 0.040359 \n",
" 0.1 0.047094 0.042053 \n",
" 0.2 0.055149 0.043369 \n",
" 0.5 0.086910 0.042562 \n",
" ham 0.0 0.041587 0.040464 \n",
" 0.1 0.058526 0.051663 \n",
" 0.2 0.079211 0.061562 \n",
" 0.5 0.292404 0.144814 "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"BATCH_IDs = ['1608030254']\n",
"DATASET_DIR = os.path.join('.')\n",
"\n",
"DF = join_repetitions(os.path.join(DATASET_DIR, '%s-df.dat' % batch_id) for batch_id in BATCH_IDs)\n",
"DF = DF.xs('trec2007-1607201347', axis=0, level='dataset')\n",
"\n",
"## reorder to put repetition at the end of the list\n",
"level_names = list(DF.index.names)\n",
"level_names.remove('repetition')\n",
"level_names.append('repetition')\n",
"level_names\n",
"DF = DF.reorder_levels(level_names).sort_index()\n",
"DF = DF.mean(level=DF.index.names[:-1]) ## assuming repetition is at the end\n",
"DF = DF.select(lambda x: x[1] in ['dictionary', 'empty', 'ham']) ## remove focussed attack\n",
"\n",
"display(DF)\n"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAABLEAAAMMCAYAAACsTDVxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nG2lEAAAAAElFTkSuQmCC\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x7f18e83b6358>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"visualise(DF, dimension_rows='classifier', dimension_color='attack')"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAABLEAAAMMCAYAAACsTDVxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAACIvP8fQX0y8AxbUTcAAAAASUVORK5CYII=\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x7f18b9451390>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"visualise(DF, dimension_rows='attack', dimension_color='classifier')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### With grey lines in background"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"attack_to_plot = 'focussed'"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr>\n",
" <th>classifier</th>\n",
" <th colspan=\"10\" halign=\"left\">adaline</th>\n",
" <th>...</th>\n",
" <th colspan=\"10\" halign=\"left\">naive bayes</th>\n",
" </tr>\n",
" <tr>\n",
" <th>repetition</th>\n",
" <th>1</th>\n",
" <th>2</th>\n",
" <th>3</th>\n",
" <th>4</th>\n",
" <th>5</th>\n",
" <th>6</th>\n",
" <th>7</th>\n",
" <th>8</th>\n",
" <th>9</th>\n",
" <th>10</th>\n",
" <th>...</th>\n",
" <th>11</th>\n",
" <th>12</th>\n",
" <th>13</th>\n",
" <th>14</th>\n",
" <th>15</th>\n",
" <th>16</th>\n",
" <th>17</th>\n",
" <th>18</th>\n",
" <th>19</th>\n",
" <th>20</th>\n",
" </tr>\n",
" <tr>\n",
" <th>% poisoned</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0.0</th>\n",
" <td>0.963328</td>\n",
" <td>0.963727</td>\n",
" <td>0.963104</td>\n",
" <td>0.963165</td>\n",
" <td>0.964223</td>\n",
" <td>0.962009</td>\n",
" <td>0.964184</td>\n",
" <td>0.963828</td>\n",
" <td>0.964080</td>\n",
" <td>0.964509</td>\n",
" <td>...</td>\n",
" <td>0.961434</td>\n",
" <td>0.962809</td>\n",
" <td>0.962947</td>\n",
" <td>0.960310</td>\n",
" <td>0.962071</td>\n",
" <td>0.961038</td>\n",
" <td>0.962954</td>\n",
" <td>0.961678</td>\n",
" <td>0.961197</td>\n",
" <td>0.962256</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.1</th>\n",
" <td>0.962858</td>\n",
" <td>0.500000</td>\n",
" <td>0.960620</td>\n",
" <td>0.962020</td>\n",
" <td>0.952392</td>\n",
" <td>0.959485</td>\n",
" <td>0.960972</td>\n",
" <td>0.955145</td>\n",
" <td>0.955459</td>\n",
" <td>0.958910</td>\n",
" <td>...</td>\n",
" <td>0.805451</td>\n",
" <td>0.954179</td>\n",
" <td>0.954124</td>\n",
" <td>0.955221</td>\n",
" <td>0.953300</td>\n",
" <td>0.956735</td>\n",
" <td>0.952085</td>\n",
" <td>0.816861</td>\n",
" <td>0.954450</td>\n",
" <td>0.955365</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.2</th>\n",
" <td>0.961498</td>\n",
" <td>0.961754</td>\n",
" <td>0.500000</td>\n",
" <td>0.963632</td>\n",
" <td>0.962396</td>\n",
" <td>0.960464</td>\n",
" <td>0.961826</td>\n",
" <td>0.961684</td>\n",
" <td>0.964699</td>\n",
" <td>0.963339</td>\n",
" <td>...</td>\n",
" <td>0.947149</td>\n",
" <td>0.942121</td>\n",
" <td>0.940075</td>\n",
" <td>0.917506</td>\n",
" <td>0.941401</td>\n",
" <td>0.944502</td>\n",
" <td>0.942193</td>\n",
" <td>0.937344</td>\n",
" <td>0.946836</td>\n",
" <td>0.934720</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.5</th>\n",
" <td>0.961498</td>\n",
" <td>0.955309</td>\n",
" <td>0.945716</td>\n",
" <td>0.500000</td>\n",
" <td>0.500000</td>\n",
" <td>0.500000</td>\n",
" <td>0.500000</td>\n",
" <td>0.951032</td>\n",
" <td>0.500000</td>\n",
" <td>0.500000</td>\n",
" <td>...</td>\n",
" <td>0.919234</td>\n",
" <td>0.885238</td>\n",
" <td>0.902595</td>\n",
" <td>0.594011</td>\n",
" <td>0.903862</td>\n",
" <td>0.863291</td>\n",
" <td>0.905100</td>\n",
" <td>0.889439</td>\n",
" <td>0.900034</td>\n",
" <td>0.913382</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>4 rows × 60 columns</p>\n",
"</div>"
],
"text/plain": [
"classifier adaline \\\n",
"repetition 1 2 3 4 5 6 \n",
"% poisoned \n",
"0.0 0.963328 0.963727 0.963104 0.963165 0.964223 0.962009 \n",
"0.1 0.962858 0.500000 0.960620 0.962020 0.952392 0.959485 \n",
"0.2 0.961498 0.961754 0.500000 0.963632 0.962396 0.960464 \n",
"0.5 0.961498 0.955309 0.945716 0.500000 0.500000 0.500000 \n",
"\n",
"classifier ... naive bayes \\\n",
"repetition 7 8 9 10 ... 11 \n",
"% poisoned ... \n",
"0.0 0.964184 0.963828 0.964080 0.964509 ... 0.961434 \n",
"0.1 0.960972 0.955145 0.955459 0.958910 ... 0.805451 \n",
"0.2 0.961826 0.961684 0.964699 0.963339 ... 0.947149 \n",
"0.5 0.500000 0.951032 0.500000 0.500000 ... 0.919234 \n",
"\n",
"classifier \\\n",
"repetition 12 13 14 15 16 17 \n",
"% poisoned \n",
"0.0 0.962809 0.962947 0.960310 0.962071 0.961038 0.962954 \n",
"0.1 0.954179 0.954124 0.955221 0.953300 0.956735 0.952085 \n",
"0.2 0.942121 0.940075 0.917506 0.941401 0.944502 0.942193 \n",
"0.5 0.885238 0.902595 0.594011 0.903862 0.863291 0.905100 \n",
"\n",
"classifier \n",
"repetition 18 19 20 \n",
"% poisoned \n",
"0.0 0.961678 0.961197 0.962256 \n",
"0.1 0.816861 0.954450 0.955365 \n",
"0.2 0.937344 0.946836 0.934720 \n",
"0.5 0.889439 0.900034 0.913382 \n",
"\n",
"[4 rows x 60 columns]"
]
},
"execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"## with grey lines in the back\n",
"df_plot_iters = df.unstack(['classifier', 'repetition']).ix[attack_to_plot]['AUC']\n",
"df_plot_iters"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th>classifier</th>\n",
" <th>adaline</th>\n",
" <th>logistic regression</th>\n",
" <th>naive bayes</th>\n",
" </tr>\n",
" <tr>\n",
" <th>% poisoned</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0.0</th>\n",
" <td>0.963954</td>\n",
" <td>0.990041</td>\n",
" <td>0.962217</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.1</th>\n",
" <td>0.914405</td>\n",
" <td>0.989954</td>\n",
" <td>0.922401</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.2</th>\n",
" <td>0.868746</td>\n",
" <td>0.989584</td>\n",
" <td>0.915868</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0.5</th>\n",
" <td>0.682402</td>\n",
" <td>0.987840</td>\n",
" <td>0.849503</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
"classifier adaline logistic regression naive bayes\n",
"% poisoned \n",
"0.0 0.963954 0.990041 0.962217\n",
"0.1 0.914405 0.989954 0.922401\n",
"0.2 0.868746 0.989584 0.915868\n",
"0.5 0.682402 0.987840 0.849503"
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df_plot = df.mean(level=df.index.names[:-1]).unstack(['classifier']).ix[attack_to_plot]\n",
"df_plot['AUC']"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAA6cAAAJoCAYAAACa1ZelAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSZIk9dyfEMyIGUP7QdcAAAAASUVORK5CYII=\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x7f52a566cb70>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig, axis = plt.subplots(1,1, figsize=(16,10))\n",
"df_plot_iters.plot(ax=axis, color='lightgray', lw=1, legend=None)\n",
"df_plot['AUC'].plot(ax=axis)\n",
"\n",
"plt.ylim((0.5, 1))\n",
"plt.title('ROC AUC')\n",
"plt.show()"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.5.2"
}
},
"nbformat": 4,
"nbformat_minor": 0
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment