Data Modeling With Graphs

Talk by Ian Robinson, Director Customer Success @NetoTechnology at Data Science London meetup

Transcript

1. Data  Modeling  With  Graphs   @ianSrobinson   ian@neotechnology.com    
2. complexity = f(size, semi-structure, connectedness) Data  Complexity  
4. Large,  Variably-­‐Structured,  Densely-­‐Connected  
5. Graph  Databases   •  Store   •  Manage   •  Query   data  
6. Property  Graph  Data  Model  
7. RelaDonships  –  Variable  Structure   Defined  with  regard  to  node  instances,   not  classes  of  nodes  
8. RelaDonships  –  Performance   Index-­‐free  adjacency  
9. Modeling  with  RelaDonships   •  Align  with  use  cases   – When  querying,  well-­‐named  relaDonships  help   discover  only  what  is  absolutely  necessary   •  No  ‘specializaDon  tax’  
10. Design  For  Queryability   MATCH (company)<-[:WORKS_FOR]-(me:Person)-[:HAS_SKILL]->(skill), (company)<-[:WORKS_FOR]-(colleague)-[:HAS_SKILL]->(skill) WHERE me.name = {name} RETURN colleague.name AS name, count(skill) AS score, collect(skill.name) AS skills ORDER BY score DESC As  an  employee     I  want  to  know  who  in  the  company  I   work  for  has  similar  skills  to  me     So  that  we  can  exchange  knowledge   (:Company)<-[:WORKS_FOR]-(:Person)-[:HAS_SKILL]->(:Skill) Person  WORKS_FOR  Company   Person  HAS_SKILL  Skill ? Which  people,  who  work  for  the  same   company  as  me,  have  similar  skills  to  me?
11. ApplicaDon/End-­‐User  Goals   As  an  employee     I  want  to  know  who  in  the  company  I   work  for  has  similar  skills  to  me     So  that  we  can  exchange  knowledge  
12. IdenDfy  EnDDes  and  ConnecDons   Which  people,  who  work  for  the  same  company   as  me,  have  similar  skills  to  me?     (:Person)-­‐[:WORKS_FOR]-­‐>(:Company),   (:Person)-­‐[:HAS_SKILL]-­‐>(:Skill)      
13. Consolidate  Paths   (:Person)-[:WORKS_FOR]->(:Company), (:Person)-[:HAS_SKILL]->(:Skill) (:Company)<-[:WORKS_FOR]-(:Person)-[:HAS_SKILL]->(:Skill)
14. Candidate  Data  Model   (:Company)<-[:WORKS_FOR]-(:Person)-[:HAS_SKILL]->(:Skill)
15. Express  QuesDon  as  Graph  Pabern   Which  people,  who  work  for  the  same  company   as  me,  have  similar  skills  to  me?  
16. Cypher  Query   Which  people,  who  work  for  the  same  company   as  me,  have  similar  skills  to  me?   MATCH (company)<-[:WORKS_FOR]-(me:Person)-[:HAS_SKILL]->(skill), (company)<-[:WORKS_FOR]-(colleague)-[:HAS_SKILL]->(skill) WHERE me.name = {name} RETURN colleague.name AS name, count(skill) AS score, collect(skill.name) AS skills ORDER BY score DESC
17. Graph  Pabern   Which  people,  who  work  for  the  same  company   as  me,  have  similar  skills  to  me?   MATCH (company)<-[:WORKS_FOR]-(me:Person)-[:HAS_SKILL]->(skill), (company)<-[:WORKS_FOR]-(colleague)-[:HAS_SKILL]->(skill) WHERE me.name = {name} RETURN colleague.name AS name, count(skill) AS score, collect(skill.name) AS skills ORDER BY score DESC
18. Which  people,  who  work  for  the  same  company   as  me,  have  similar  skills  to  me?   MATCH (company)<-[:WORKS_FOR]-(me:Person)-[:HAS_SKILL]->(skill), (company)<-[:WORKS_FOR]-(colleague)-[:HAS_SKILL]->(skill) WHERE me.name = {name} RETURN colleague.name AS name, count(skill) AS score, collect(skill.name) AS skills ORDER BY score DESC Anchor  Pabern  in  Graph   Search  nodes  labeled   Person,  matching  on   name  property  
19. Create  ProjecDon  of  Results   Which  people,  who  work  for  the  same  company   as  me,  have  similar  skills  to  me?   MATCH (company)<-[:WORKS_FOR]-(me:Person)-[:HAS_SKILL]->(skill), (company)<-[:WORKS_FOR]-(colleague)-[:HAS_SKILL]->(skill) WHERE me.name = {name} RETURN colleague.name AS name, count(skill) AS score, collect(skill.name) AS skills ORDER BY score DESC
20. First  Match  
21. Second  Match  
22. Third  Match  
23. Running  the  Query   +-----------------------------------+ | name | score | skills | +-----------------------------------+ | "Lucy" | 2 | ["Java","Neo4j"] | | "Bill" | 1 | ["Neo4j"] | +-----------------------------------+ 2 rows
24. Thank  you   @ianSrobinson   ian@neotechnology.com