Graphs in Action: In-depth look at Neo4j in Production
•
1 like•481 views
View the slides from 'Graphs in Action' by William Lyon on the Neo4j Developer Relations team presented at GraphTalk Denver.
Recommended
Recommended
More Related Content
What's hot
What's hot (20)
Similar to Graphs in Action: In-depth look at Neo4j in Production
Similar to Graphs in Action: In-depth look at Neo4j in Production (20)
More from Neo4j
More from Neo4j (20)
Recently uploaded
Recently uploaded (20)
Graphs in Action: In-depth look at Neo4j in Production
- 1. 3:30 - 4:00 4:00 - 4:30 4:30 - 5:00 5:00 - 7:00 The Connected Data Imperative Graphs In Action Customer Use Cases Social Time Agenda May 2017 Denver
- 2. Graphs In Action Neo4j Graph Talk - Denver 2017 William Lyon @lyonwj
- 3. William Lyon Developer Relations Engineer @neo4j will@neo4j.com @lyonwj lyonwj.com
- 4. Agenda • Graph database - developer perspective • (relational)-[:TO]->(graph) • Use cases • Fraud Detection • Knowledge Graphs • Recommendations • Developer tooling
- 5. Neo4j Graph Database • Property graph data model • (open)Cypher query language • Native graph processing • Language drivers • Open source neo4j.com
- 7. Labeled Property Graph Model
- 8. Labeled Property Graph Model
- 16. Cypher
- 19. FROM RDBMS TO GRAPHS
- 21. Northwind
- 22. Northwind - the canonical RDBMS Example
- 26. Starting with the ER Diagram
- 27. Locate the Foreign Keys
- 28. Drop the Foreign Keys
- 29. Find the JOIN Tables
- 30. (Simple) JOIN Tables Become Relationships
- 31. Attributed JOIN Tables -> Relationships with Properties
- 32. As a Graph
- 34. Northwind
- 35. Northwind
- 45. POWERING AN APP
- 46. Simple App
- 47. Simple App
- 53. Case Studies for Knowledge Graphs, Fraud Detection, and Recommendation Engines Neo4j Case Studies
- 54. Real-Time Recommendations Dynamic Pricing Artificial Intelligence & IoT-applications Fraud Detection Network Management Customer Engagement Supply Chain Efficiency Identity and Access Management Relationship-Driven Applications
- 55. Shopping Recommendations Examples of companies that use Neo4j, the world’s leading graph database, for recommendation and personalization engines. Adidas uses Neo4j to combine content and product data into a single, searchable graph database which is used to create a personalized customer experience “We have many different silos, many different data domains, and in order to make sense out of our data, we needed to bring those together and make them useful for us,” – Sokratis Kartelias, Adidas eBay ShopBot Personal Shopping Companion in FB Messenger “ShopBot uses its Knowledge Graph to understand user requests and generate follow-up questions to refine requests before searching for the items in eBay’s inventory. In a search query for “bags” for example, purple nodes represent “categories,” green “attributes” and pink are “values” for those attributes.” – RJ Pittman Blog, eBay Walmart uses Neo4j to give customer best web experience through relevant and personal recommendations “As the current market leader in graph databases, and with enterprise features for scalability and availability, Neo4j is the right choice to meet our demands”. - Marcos Vada, Walmart Product recommendations Personalization Linkedin Chitu seeks to engage Chinese jobseekers through a game-like user interface that is available on both desktop and mobile devices. “The challenge was speed,” said Dong Bin, Manager of Development at Chitu. “Due to the rate of growth we saw from our competitors in the Chinese market, we knew that we had to launch Chitu as quickly as possible.” Social Network Additional Case Studies
- 56. Case Study: Knowledge Graphs at eBay
- 57. Case Study: Knowledge Graphs at eBay
- 58. Case Study: Knowledge Graphs at eBay
- 59. Case Study: Knowledge Graphs at eBay
- 60. Bags Case Study: Knowledge Graphs at eBay
- 61. Men’s Backpack Handbag Case Study: Knowledge Graphs at eBay
- 62. https://shopbot.ebay.com/ Try it out at: Case Study: Knowledge Graphs at eBay
- 64. Organized in groups Synthetic Identities Stolen Identities Hijacked Devices Who Are Today’s Fraudsters?
- 65. Fraud Rings and Synthetic Identities
- 66. ACCOUNT HOLDER 2 Modeling a fraud ring as a graph ACCOUNT HOLDER 1 ACCOUNT HOLDER 3
- 67. ACCOUNT HOLDER 2 ACCOUNT HOLDER 1 ACCOUNT HOLDER 3 CREDIT CARD BANK ACCOUNT BANK ACCOUNT BANK ACCOUNT PHONE NUMBER UNSECURED LOAN SSN 2 UNSECURED LOAN Modeling a fraud ring as a graph
- 68. ACCOUNT HOLDER 2 ACCOUNT HOLDER 1 ACCOUNT HOLDER 3 CREDIT CARD BANK ACCOUNT BANK ACCOUNT BANK ACCOUNT ADDRESS PHONE NUMBER PHONE NUMBER SSN 2 UNSECURED LOAN SSN 2 UNSECURED LOAN Modeling a fraud ring as a graph
- 75. Real-time Package Routing • Large postal service with over 500k employees • Neo4j routes 7M+ packages daily at peak, with peaks of 5,000+ routing operations per second. Real-time promotion recommendations • Record “Cyber Monday” sales • About 35M daily transactions • Each transaction is 3-22 hops • Queries executed in 4ms or less • Replaced IBM Websphere commerce Real-time pricing engine • 300M pricing operations per day • 10x transaction throughput on half the hardware compared to Oracle • Presentation at http:// graphconnect.com/gc2016-sf/ • Replaced Oracle database Recommendations, Pricing and Routing
- 76. Recommendations drive user engagement
- 77. “35 percent of what consumers purchase on Amazon and 75 percent of what they watch on Netflix come from product recommendations” http://www.mckinsey.com/industries/retail/our-insights/how-retailers-can-keep-up-with-consumers
- 78. Product Recommendations Effective product recommendation algorithms has become the new standard in online retail — directly affecting revenue streams and the shopping experience. Logistics/Delivery Routing recommendations allows companies to save money on routing and delivery, and provide better and faster service. Promotion recommendations Building powerful personalized promotion engines is another area within retail that requires input from multiple data sources, and real-time, session based queries, which is an ideal task to solve with Neo4j. Today Recommendation Engines are At the Core of Digitization in Retail
- 79. Powerful recommendation engines rely on the connections between multiple sources of data
- 80. Powerful recommendation engines rely on the connections between multiple sources of data
- 81. Mobile Brick & Mortar Multi-Channel Web Product Recommendations
- 82. Mobile Brick & Mortar Web
- 83. Personalized Promotions Personalized Real-Time Recommendations Personalized Real-Time Recommendations
- 84. Data-Model (Expressed as a graph) Categor y Categor y Produc t Produc t Produc t Collaborative Filtering An algorithm that considers users interactions with products, with the assumption that other users will behave in similar ways. Algorithm Types Content Based An algorithm that considers similarities between products and categories of products. Custome r Custome r Produc t Produc t Produc t
- 85. Category Price ConfigurationsLocation Silos & Polyglot Persistence Purchase ViewReviewReturn In-store PurchasesInventory Products Customers / Users Location Purchases RELATIONAL DB WIDE COLUMN STORE Views DOCUMENT STORE User Review RELATIONAL DB In-Store Purchase Shopping Cart KEY VALUE STORE Product Catalogue DOCUMENT STORE
- 86. Purchases RELATIONAL DB WIDE COLUMN STORE Views DOCUMENT STORE User Review RELATIONAL DB In-Store Purchase Shopping Cart KEY VALUE STORE Product Catalogue DOCUMENT STORE Silos & Polyglot Persistence Category Price ConfigurationsLocation Purchase ViewReviewReturn In-store PurchasesInventory Products Customers / Users Location
- 87. Purchases RELATIONAL DB WIDE COLUMN STORE Views DOCUMENT STORE User Review RELATIONAL DB In-Store Purchase Shopping Cart KEY VALUE STORE Product Catalogue DOCUMENT STORE Category Price ConfigurationsLocation Polyglot Persistence Purchase ViewReviewReturn In-store PurchasesInventory LocationCategory Price ConfigurationsLocation Purchase ViewReviewReturn In-store PurchasesInventory Products Customers / Users Location
- 88. Data Lake Purchases RELATIONAL DB Product Catalogue DOCUMENT STORE WIDE COLUMN STORE Views DOCUMENT STORE User Review RELATIONAL DB In-Store Purchase Shopping Cart KEY VALUE STORE Recommendations require an operational workload — it’s in the moment, real-time! Good for Analytics, BI, Map Reduce Non-Operational, Slow Queries
- 89. Purchases RELATIONAL DB Product Catalogue DOCUMENT STORE WIDE COLUMN STORE Views DOCUMENT STORE User Review RELATIONAL DB In-Store Purchase Shopping Cart KEY VALUE STORE Connector Drivers: Java | JavaScript | Python | .Net | PHP | Go | Ruby Apps and Systems Real-Time Queries
- 90. Using Data Relationships for Recommendations Content-based filtering Recommend items based on what users have liked in the past Collaborative filtering Predict what users like based on the similarity of their behaviors, activities and preferences to others Movie Person Person RATED SIMILARITY rating: 7 value: .92
- 95. In Cypher
- 96. In Cypher Basic initial approach. Improvements: - aggregate across all purchases - scoring / normalize - compute similarity metrics
- 99. Content Filtering w/ Cypher
- 100. Content Filtering - Concept Hierarchy
- 101. Content Filtering - Concept Hierarchy
- 102. Content Filtering - Concept Hierarchy
- 103. Content Filtering - Concept Hierarchy w/ Cypher
- 104. Content Filtering - Concept Hierarchy w/ Cypher Basic initial approach. Improvements: - aggregate across all purchases - cold start - variable length concept hierarchy - tag similarity / clusters
- 106. Want to learn more? graphdatabases.com neo4j.com/sandbox