Skip to content

Instantly share code, notes, and snippets.

@satish3485

satish3485/satish kumar

Last active May 10, 2016 16:22
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save satish3485/5ce3fecca9069c7f5ef239601aa8fdc5 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save satish3485/5ce3fecca9069c7f5ef239601aa8fdc5 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
satish kumar
= REPLACEME: TITLE OF YOUR GRAPHGIST
:neo4j-version: 2.3.0
:author: SATISH KUMAR
= Homework Assignement: Create a small Graph Use Case Model with Neo4j and Cypher
== Domain and Data
Take a domain that you are really interested in, be it politics, beer, music, arts, cooking, sports, travel, software, organizations.
Draw a model of the domain on a whiteboard or digitally, add a few example entities (nodes) you know and their relationships.
Think about at least 3 questions you want to answer using this _connected_ data.
Remember to add all the information you need to answer those questions.
You should end up with *at least 15 nodes and 20 relationships*.
== GraphGist Format
Now go to http://portal.graphgist.org/about and learn about the GraphGist documentation format (which is basically AsciiDoc a text-markup language with a few embellishments).
You can either start from the template that is provided there or a blank plain text file.
You can also explore the other GraphGists on the portal for further inspiration.
== Setup
Add a snappy *title* for your demo and *some prose* that describes the domain and the types of its entities and their relationships.
Now *add an image* link, e.g. to the drawing you created earlier.
Now you can start creating your domain with Cypher, Neo4j's query language in a *Setup* section.
Follow the syntax example in the template or on the portal page.
image::http://cdn.yourarticlelibrary.com/wp-content/uploads/2013/12/415-1.jpg[width=200]
== Setup
////
A Cypher query to setup the database
Please use a small sample of your domain, at most 150 nodes and 200 relationships are enough for the pedagocial example.
You can link to the setup of a larger dataset or LOAD CSV queries in a second file of your GitHub Gist at the end.
In your setup query you can also use LOAD CSV loading CSV files from your GitHub Gist, like here: https://gist.github.com/jexp/f78df7b232d0faa171ff
////
//setup
//hide
[source,cypher]
----
CREATE
(a:Person {name: 'Mahendra Singh Dhoni'}),
(b:Woman {name: 'Virat Kohli'}),
(c:Child {name: 'Ajinkya Rahane'}),
(d:Bigger {name: 'Shikhar Dhawan'}),
(e:Smallest {name: 'Rohit Sharma'}),
(f:Person {name: 'Stuart Binny'}),
(g:Person {name: 'Suresh Raina'}),
(h:Person {name: 'Ravindra Jadeja'}),
(i:Person {name: 'Ambati Rayudu'}),
(j:Person {name: 'Axar Patel'}),
(k:Person {name: 'Ravichandran Ashwin'}),
(l:Person {name: 'Bhuvneshwar Kumar'}),
(m:Person {name: 'Mohammed Shami'}),
(n:Person {name: 'Umesh Yadav'}),
(o:Person {name: 'Ishant Sharma'}),
(p:Dog {name: 'Tommy'}),
(a)-[:HUSBAND]->(b),
(a)-[:SON]->(e),
(b)-[:SON]->(e),
(b)-[:FRIENDS_WITH]->(c),
(a)-[:FRIENDS_WITH]->(d),
(e)-[:DAUGHTER]->(f),
(e)-[:SON]->(g),
(h)-[:WIFE]->(e),
(h)-[:DAUGHTER]->(f),
(h)-[:SON]->(g),
(b)-[:AUNT]->(i),
(a)-[:UNCLE]->(i),
(j)-[:GRAND_FATHER]->(f),
(k)-[:GRAND_MOTHER]->(g),
(l)-[:SISTER_IN_LAW]->(k),
(m)-[:BROTHER_IN_LAW]->(l),
(n)-[:GIRLFRIEND]->(k),
(o)-[:TEACHER]->(l),
(a)-[:LOVES]->(p),
(b)-[:LOVES]->(p),
(c)-[:LOVES]->(p),
(d)-[:LOVES]->(p),
(f)-[:LOVES]->(p),
(g)-[:LOVES]->(p),
(h)-[:LOVES]->(p),
(i)-[:LOVES]->(p),
(j)-[:LOVES]->(p),
(k)-[:LOVES]->(p),
(h)-[:SON]->(g),
(i)-[:PROF]->(o),
(e)-[:BROTHER]->(a),
(j)-[:FINANCE_ADVICER]->(f)
----
//graph
== Name all the people who loves dogs
// REMOVEME: Describe what this use-case is solving and how the query represents that
// REMOVEME: A Cypher query with tabular output, returning scalar values
// Name all the people who loves dogs
[source,cypher]
----
//MATCH (a:Person {name: 'Mahendra Singh Dhoni'})-[:HUSBAND]-(:Woman)-[:LOVES]-(name:Dog)
MATCH (a:Person)-[:LOVES]-(name:Dog)
RETURN a.name as name,count(*) as count
----
//table
== REPLACEME: SECOND USE-CASE TITLE
// REMOVEME: Describe what this use-case is solving and how the query represents that
// REMOVEME: A Cypher query to with graph output
[source,cypher]
----
MATCH path=(a:Person {name: 'Bhuvneshwar Kumar'})-[:SISTER_IN_LAW]-(:Person)-[:GRAND_MOTHER]-(fof:Person)
RETURN path, a, fof
----
//graph_result
Render the results either as `//table` *or* `//graph_result`.
== Submission & Grading
While you're working and when you are done, you can use the "Preview" functionality of the GraphGist portal to render your document in your browser.
Go to http://portal.graphgist.org/submit_graphgist add the public URL of your file, your title and click `Preview`.
When done, either hand in the preview link or submit your link to the portal and share the resulting link with your lecturer.
*The grading will be on completeness, correct usage of Cypher, good prose and educational quality of your queries (i.e. at least 2 of the 3 should use relationships or paths).*
== Resources
* http://neo4j.com/graphacademy/online-course-getting-started/[Online Course]
* http://neo4j.com/developer/cypher[Cypher Information on the Developer Pages]
* http://neo4j.com/docs/developer-manual/current/#cypher-getting-started[Cypher in the Neo4j Manual]
* http://neo4j.com/docs/cypher-refcard/current/[Cypher Refcard]
* http://portal.graphgist.org/about[GraphGist Howto]
* https://gist.github.com/jexp/a56f7e7587ce5ed97561[Template]
Created by {author}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment