Holger Spill: An introduction to Python and graph databases with Neo4j

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Holger Spill:
An introduction to Python and graph databases with Neo4j
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
@ Kiwi PyCon 2014 - Sunday, 14 Sep 2014 - Track 2
http://kiwi.pycon.org/

**Audience level**

Novice

**Description**

A number of organisations have realised that the relational model for databases is insufficient for certain scenarios. Graph databases are good at modelling relationships. This talk will introduce you to Neo4j (a graph database engine) with Python. How to set it up, how to use it and some questions you can answer with it.

**Abstract**

This talk will introduce you to Neo4j (a graph database engine) with Python. How to set it up, how to use it and some questions you can answer with it.

You don’t need any previous experience with Neo4j or NoSQL databases and very little experience with Python.

We'll cover: Introduction to Graphs Setting up your toolbox Neo4j and Cypher A simple real-world example use case with Neo4j

**YouTube**

https://www.youtube.com/watch?v=ps5RtmWKwLY

Transcript

1. AN INTRODUCTION TO PYTHON AND
   GRAPH DATABASES WITH NEO4J
   Holger Spill
   Kiwi PyCon – 14 September 2014
   

   View Slide

2. http://lorenabarba.com/blog/
   

   View Slide

3. A GRAPH
   Node
   Relationship
   

   View Slide

4. https://wiki.cs.umd.edu/cmsc734_09/index.php
   David Bowie Defines Rock Royalty
   ANOTHER GRAPH
   

   View Slide

5. WHERE ARE GRAPHS USEFUL?
    Social Network
    Network Impact Analysis
    Route Finding
    Recommendations
    Logistics
    Access Control
    Fraud Analysis
    Securities and Debt
    …
   

   View Slide

6. THE PROPERTIES GRAPH
   language:’Swedish’
   open_source:true
   year:2007
   by Nicole White - @_nicolemargaret
   

   View Slide

7. FROM SQL TO CYPHER
   

   View Slide

8. EXAMPLE: YOUTUBE IN SQL
   ID Name
   1 Alice
   2 Bob
   3 Charles
   4 David
   ID Name
   1 Bob’s Gaming Channel
   2 Bob’s Cute Dog
   3 Cooking with Charles
   4 David’s How-To Channel
   5 Disco Dancing with David
   User
   ID
   Channel
   ID
   2 1
   2 2
   3 3
   4 4
   4 5
   USERS
   CHANNELS
   USERS_CHANNELS
   User ID Channel ID
   1 3
   1 4
   2 3
   2 5
   3 1
   USERS_SUBSCRIPTIONS
   

   View Slide

9. EXAMPLE: YOUTUBE IN A GRAPH
   name:’Alice’
   name:’Bob’
   name:’Charles’
   name:’
   David’
   name:’Cooking
   with Charles’
   name:’David’s How-
   To Channel’
   name:’Disco Dancing
   with David’
   name:’Bob’s Gaming
   Channel’
   name:’Bob’s Cute
   Dog’
   User
   Channel
   

   View Slide

10. CYPHER QUERY LANGUAGE
    .|'''', '||
    | ||
    || '|| ||` '||''|, ||''|, .|''|, '||''|
    || `|..|| || || || || ||..|| ||
    `|....' || ||..|' .|| || `|... .||.
    , |' ||
    '' .||
    

    View Slide

11. EXAMPLE: YOUTUBE IN A GRAPH
    name:’Alice’
    name:’Bob’
    name:’Charles’
    name:’
    David’
    name:’Cooking
    with Charles’
    name:’David’s How-
    To Channel’
    name:’Disco Dancing
    with David’
    name:’Bob’s Gaming
    Channel’
    name:’Bob’s Cute
    Dog’
    User
    Channel
    MATCH (c:Channel) RETURN c;
    

    View Slide

12. EXAMPLE: YOUTUBE IN A GRAPH
    name:’Alice’
    name:’Bob’
    name:’Charles’
    name:’
    David’
    name:’Cooking
    with Charles’
    name:’David’s How-
    To Channel’
    name:’Disco Dancing
    with David’
    name:’Bob’s Gaming
    Channel’
    name:’Bob’s Cute
    Dog’
    User
    Channel
    MATCH (u:User) RETURN u;
    

    View Slide

13. EXAMPLE: YOUTUBE IN A GRAPH
    name:’Alice’
    name:’Bob’
    name:’Charles’
    name:’
    David’
    name:’Cooking
    with Charles’
    name:’David’s How-
    To Channel’
    name:’Disco Dancing
    with David’
    name:’Bob’s Gaming
    Channel’
    name:’Bob’s Cute
    Dog’
    User
    Channel
    MATCH (bob:User {name:’Bob’}) RETURN bob;
    

    View Slide

14. EXAMPLE: YOUTUBE IN A GRAPH
    name:’Alice’
    name:’Bob’
    name:’Charles’
    name:’
    David’
    name:’Cooking
    with Charles’
    name:’David’s How-
    To Channel’
    name:’Disco Dancing
    with David’
    name:’Bob’s Gaming
    Channel’
    name:’Bob’s Cute
    Dog’
    User
    Channel
    MATCH (u:User)-[r:OPERATES]->(c:Channel) RETURN u,r,c;
    

    View Slide

15. EXAMPLE: YOUTUBE IN A GRAPH
    name:’Alice’
    name:’Bob’
    name:’Charles’
    name:’
    David’
    name:’Cooking
    with Charles’
    name:’David’s How-
    To Channel’
    name:’Disco Dancing
    with David’
    name:’Bob’s Gaming
    Channel’
    name:’Bob’s Cute
    Dog’
    User
    Channel
    MATCH p=(:User)-[:OPERATES]->(:Channel) RETURN p;
    

    View Slide

16. EXAMPLE: YOUTUBE IN A GRAPH
    name:’Alice’
    name:’Bob’
    name:’Charles’
    name:’
    David’
    name:’Cooking
    with Charles’
    name:’David’s How-
    To Channel’
    name:’Disco Dancing
    with David’
    name:’Bob’s Gaming
    Channel’
    name:’Bob’s Cute
    Dog’
    User
    Channel MATCH p=(:User {name:'Charles'})-[*1..4]-(:User {name:'David'})
    RETURN p LIMIT 1;
    

    View Slide

17. CYPHER QUERY STRUCTURE
    MATCH (n:Label)-[:REL]->(m:Label)
    WHERE n.prop < 42
    WITH n, count(m) as cnt,
    collect(m.attr) as attrs
    WHERE cnt > 12
    RETURN n.prop,
    extract(a2 in
    filter(a1 in attrs
    WHERE a1 =~ "...-.*")
    | substr(a2,4,size(a2)-1)]
    AS ids
    ORDER BY length(ids) DESC
    LIMIT 10
    

    View Slide

18. PROS & CONS
    Strengths
     Powerful data model, as general as RDBMS
     Whiteboard friendly, flexible development
     Fast, for connected data
     Easy to query
    Weaknesses:
     Sharding
     Global Queries and Number Crunching
     Binary Data / Blobs
     Requires conceptual shift
    

    View Slide

19. READY, STEADY, GO START
    … or how to install Neo4j on Ubuntu
     Start with a stock Ubuntu install
     Install Oracle JDK7 (see askubuntu)
     Download Neo4j Linux Community Edition
     Follow the Neo4j installation instructions for Linux
    cd ; bin/neo4j start
     Visit http://localhost:7474/browser/
    

    View Slide

20. READY, STEADY, GO START
    

    View Slide

21. MAKING FRIENDS
    www.neo4j.org
    www.python.org
    

    View Slide

22. PY2NEO
     simple and pragmatic Python library
     provides access via RESTful web services
     no external dependencies
     installation is straightforward
     getting started is easy
     actively maintained on GitHub by Nigel Small
    from py2neo import neo4j
    db_uri = "http://localhost:7474/db/data/"
    graph = neo4j.GraphDatabaseService(db_uri)
    neo4j.CypherQuery(graph, "Cypher goes here").run()
    

    View Slide

23. … AND ALTERNATIVES
    Object Graph Mapper (OGM)
    built on top of py2neo
    Neo4j Rest Client
    Object Graph Mapper (OGM) for Django.
    Use familiar Django models and queries
    against Neo4j.
    Python persistence framework
    for graph databases
    

    View Slide

24. REST API
    :GET /dbdata
    {
    "extensions": {},
    "node": "http://localhost:7474/db/data/node",
    "node_index": "http://localhost:7474/db/data/index/node",
    […]
    "http://localhost:7474/db/data/schema/constraint",
    "transaction": http://localhost:7474/db/data/transaction",
    "node_labels": "http://localhost:7474/db/data/labels",
    "neo4j_version": "2.1.4”
    }
    

    View Slide

25. CURRENT RELEASE
    Neo4j 2.1.4
    4 September, 2014
    Community and Enterprise Edition
     Import CSV Files with Cypher
     Dense nodes support
     New Cypher functionality
     Official support for OpenJDK 7
     ... and more
    http://neo4j.com/release-notes/neo4j-2-1-4/
    

    View Slide

26. JOIN THE CONVERSATION
    bit.ly/neo4j-google
    @neo4j
    #neo4j
    www.stackoverflow.com neo4j.meetup.com
    >>> Neo4j version, programming language, driver name, query language <<<
    

    View Slide

27. View Slide