GraphSummit Toronto: Leveraging Graphs for AI and ML
0 likes106 views
This document discusses how graph data science can be used as a secret ingredient for relationship-driven AI. It explains that traditional machine learning ignores network structure, but graph databases can store and retrieve relationships to make AI more contextual. Graph algorithms and embeddings can infer relationships and enrich data. The document provides examples of how knowledge graphs can be used for applications like recommendations, fraud detection, and knowledge management. It also outlines the key components of graph data science including graph algorithms, machine learning workflows, and the Neo4j graph database platform.
![Neo4j, Inc. All rights reserved 2022
Graph Feature Engineering
23
Human-crafted query, human-readable result
MATCH (p1:Person)-[:ENEMY]->(:Person)<-[:ENEMY]-(p2:PERSON)
MERGE (p1)-[:FRIEND]->(p2)
AI-learned formula, machine-readable result
Predefined formula, human-readable result
PageRank(Emil) = 13.25
PageRank(Amy) = 4.83
PageRank(Alicia) = 4.75
Node2Vec(Emil) =[5.4 5.1 2.4 4.5 3.1]
Node2Vec(Amy) =[2.8 1.8 7.2 0.9 3.0]
Node2Vec(Alicia)=[1.4 5.2 4.4 3.9 3.2]
Queries
Algorithms
Embeddings
Machine
Learning
Workflows
Train ML models
based on results](https://image.slidesharecdn.com/leveraginggraphsforaiandml-phanidathar-graphsummittoronto-nov1-221107194137-df2d97ed/85/GraphSummit-Toronto-Leveraging-Graphs-for-AI-and-ML-23-320.jpg)
GraphSummit Toronto: Leveraging Graphs for AI and ML
- 1. © 2022 Neo4j, Inc. All rights reserved. © 2022 Neo4j, Inc. All rights reserved. 1 Phani Dathar, PhD Data Science Solution Architect, Neo4j Graph Data Science: The Secret Ingredient for Relationship-Driven AI
- 2. Neo4j, Inc. All rights reserved 2022 Networks of People Transaction Networks Bought B ou gh t V i e w e d R e t u r n e d Bought Knowledge Networks Pl ay s Lives_in In_sport Likes F a n _ o f Plays_for Risk management, Supply chain, Orders, Payments, etc. Employees, Customers, Suppliers, Partners, Influencers, etc. Enterprise content, Domain specific content, eCommerce content, etc K n o w s Knows Knows K n o w s 2 Everything is Naturally Connected
- 3. Neo4j, Inc. All rights reserved 2022 Relationship-Driven AI ● Traditional ML ignore network structure because it’s difficult to extract ● Use the right data structures to store and retrieve relationships ● Add relationships to AI/ML pipelines to make them contextual and to unlock otherwise unattainable predictions 3 Machine Learning Pipeline
- 4. Neo4j, Inc. All rights reserved 2022 4 Let’s Start with an Example…
- 5. Neo4j, Inc. All rights reserved 2022 5 Graph to Tabular Data for ML.. Lamp Lightbulbs Pillow **HE Light bulbs Mingo X Jane X X Aditi X Fabien … …
- 6. Neo4j, Inc. All rights reserved 2022 6 What’s the Problem? Generating personalized recommendations is hard due to high dimensionality and sparse data sets ● History for every customer to generate personalized recommendations ○ Increases the problem of **sparse** and insufficient information ● Reduce dimensionality by matrix factorization and content (word) embeddings ○ Only suited for content based recommendations ● Macro level insights for cold start problem ○ Generates poor recommendations
- 7. Neo4j, Inc. All rights reserved 2022 Take advantage of ● graphs to capture the network structure ● graph queries to store and retrieve relationships ● graph algorithms to infer relationships What Should We Do?
- 8. Neo4j, Inc. All rights reserved 2022 Knowledge Graphs Graph Feature Engineering and Graph ML Graph Analytics, Investigations and Counterfactuals Integrations and Knowledge Graphs for Heuristic AI Capitalize Analysis Data Modeling 8 Graphs Enrich all Stages of AI Ecosystem
- 9. Neo4j, Inc. All rights reserved 2022 9 Queries Find the patterns you know exist. Machine Learning Uncover trends and make predictions Visualization Explore, collaborate, and explain Graph Data Science Analytics Feature Engineering Data Exploration Graph Data Science Queries Machine Learning Visualization
- 10. Neo4j, Inc. All rights reserved 2022 Predictive Maintenance Churn Prediction Fraud Detection Life Sciences Personalized Recommendations Cybersecurity Disambiguation & Segmentation Search & Master Data Mgmt. Graphs Data Science Applications
- 11. Neo4j, Inc. All rights reserved 2022 11 Graph Data Science Knowledge Graphs Graph Algorithms Graph Native Machine Learning Find the patterns you’re looking for in connected data Use unsupervised machine learning techniques to identify associations. Use embeddings to learn the features in your graph that you don’t even know are important yet. Train supervise ML models to predict links, labels, and missing data.
- 12. Neo4j, Inc. All rights reserved 2022 12 Node Represents an entity in the graph Relationship Connect nodes to each other Property Describes a node or relationship: e.g. name, age, weight etc Knowledge Graph - Building Blocks MICA ANDRE Name: “Andre” Born: May 29, 1970 Twitter: “@dan” Name: “Mica” Born: Dec 5, 1975 CAR Brand “Volvo” Model: “V70” Since: Jan 10, 2011 LOVES LOVES LOVES LIVES WITH O W N S D R I V E S
- 13. Neo4j, Inc. All rights reserved 2022 13 Database, Query Language and Visualization DATA GRAPH QUERIES GRAPH VISUALIZATION CYPHER https://neo4j.com/developer/cypher/
- 14. Neo4j, Inc. All rights reserved 2022 What can you do with a Knowledge Graph? Collaborative filtering: users who bought X, also bought Y (open- ended pattern matching) What items make you more likely to buy additional items in subsequent transactions? Traverse hierarchies - what items are similar 4+ hops out? How many flagged accounts are in the applicant’s network 4+ hops out? How many login / account variables in common? Add these metrics to your approval process What completes the connections from genes to diseases to targets? What genes can be reached 4+ hops out from a known drug target? What mechanisms in common are there between two drugs? Financial Domain Life Sciences Marketing and Recommendations
- 15. Neo4j, Inc. All rights reserved 2022 Supply Chain: Organizational Knowledge Graph VENDORS AND SUPPLIERS OPERATIONS LOGISTICS SALES & MARKETING Bill Of Materials Supply Chain Customer 360
- 16. Neo4j, Inc. All rights reserved 2022 16 Graph Data Science Knowledge Graphs Graph Algorithms Graph Native Machine Learning Find the patterns you’re looking for in connected data Use unsupervised machine learning techniques to identify associations. Use embeddings to learn the features in your graph that you don’t even know are important yet. Train supervise ML models to predict links, labels, and missing data.
- 17. Neo4j, Inc. All rights reserved 2022 17 Graph Algorithms Pathfinding & Search • Shortest Path • Single-Source Shortest Path • All Pairs Shortest Path • A* Shortest Path • Yen’s K Shortest Path • Minimum Weight Spanning Tree • K-Spanning Tree (MST) • Random Walk • Breadth & Depth First Search Centrality & Importance • Degree Centrality • Closeness Centrality • Harmonic Centrality • Betweenness Centrality & Approx. • PageRank • Personalized PageRank • ArticleRank • Eigenvector Centrality • Hyperlink Induced Topic Search (HITS) • Influence Maximization (Greedy, CELF) Community Detection • Triangle Count • K-Means • Local Clustering Coefficient • Connected Components (Union Find) • Strongly Connected Components • Label Propagation • Louvain Modularity • K-1 Coloring • Modularity Optimization • Speaker Listener Label Propagation Supervised Machine Learning • Node Classification • Link Prediction • Node Regression … and more! Heuristic Link Prediction • Adamic Adar • Common Neighbors • Preferential Attachment • Resource Allocations • Same Community • Total Neighbors Similarity • Node Similarity • K-Nearest Neighbors (KNN) • Jaccard Similarity • Cosine Similarity • Pearson Similarity • Euclidean Distance • Approximate Nearest Neighbors (ANN) Graph Embeddings • Node2Vec • FastRP • GraphSAGE • Synthetic Graph Generation • Scale Properties • Collapse Paths • One Hot Encoding • Split Relationships • Graph Export • Pregel API (write your own algos)
- 18. Neo4j, Inc. All rights reserved 2022 18 What are Graph Algorithms?
- 19. Neo4j, Inc. All rights reserved 2022 19 Enriched Knowledge Graphs Structured Unstructured Ontologies Graph Algorithms and Graph Queries Semantics, Derived relationships and additional context Natural relationships
- 20. Neo4j, Inc. All rights reserved 2022 Graph algorithms and graph embeddings are used for generating context and resolving identities/entities Identity Management / Entity Resolution Neo4j APOC Capture relationships between entities across data sources using a knowledge graph Create additional weighted relationships based on similar text description and/or other similar metadata Construct node embeddings and resolve entities based on weighted pairwise similarity between various entities Identify communities of entities based on distance between node embeddings
- 21. Neo4j, Inc. All rights reserved 2022 Personalized Recommendations Graph algorithms and graph embeddings are used for generating product recommendations and improving search relevance Capture customer interactions and customer journey using a knowledge graph Analyze customer interactions using graph queries and find customer communities based on common purchase behavior Construct node embeddings and resolve entities based on weighted pairwise similarity between various entities Generate product recommendations based on correlations between products, search queries and historical purchases
- 22. Neo4j, Inc. All rights reserved 2022 22 Graph Data Science Knowledge Graphs Graph Algorithms Graph Native Machine Learning Find the patterns you’re looking for in connected data Use unsupervised machine learning techniques to identify associations. Use graph features to learn the features in your graph that you don’t even know are important yet. Train supervise ML models to predict links, labels, and missing data.
- 23. Neo4j, Inc. All rights reserved 2022 Graph Feature Engineering 23 Human-crafted query, human-readable result MATCH (p1:Person)-[:ENEMY]->(:Person)<-[:ENEMY]-(p2:PERSON) MERGE (p1)-[:FRIEND]->(p2) AI-learned formula, machine-readable result Predefined formula, human-readable result PageRank(Emil) = 13.25 PageRank(Amy) = 4.83 PageRank(Alicia) = 4.75 Node2Vec(Emil) =[5.4 5.1 2.4 4.5 3.1] Node2Vec(Amy) =[2.8 1.8 7.2 0.9 3.0] Node2Vec(Alicia)=[1.4 5.2 4.4 3.9 3.2] Queries Algorithms Embeddings Machine Learning Workflows Train ML models based on results
- 24. Neo4j, Inc. All rights reserved 2022 24 Graph Machine Learning Graph-Native Feature Engineering Train Predictive Model Queries Algorithms Embeddings 1. Model Type 2. Property Selection 3. Train & Test 4. Model Selection Apply Model to Existing / New Data Use Predictions for Decisions Use Predictions to Enhance the Graph Publish & Share Store Model in Database The Only Completely In-Graph, ML Workflow
- 25. Neo4j, Inc. All rights reserved 2022 25 In-Graph Machine Learning Node classification: “What kind of node is this?” Link prediction: “Should there be a relationship between these nodes?” Labeled data: Pairs of nodes that are either linked or not Features: Pre-existing attributes, algorithms (pageRank), embedding
- 26. Neo4j, Inc. All rights reserved 2022 Neo4j Graph Data Science Framework Neo4j Graph Data Science Library Neo4j Database Neo4j Bloom Scalable Graph Algorithms & Analytics Workspace Native Graph Creation & Persistence Visual Graph Exploration & Prototyping
- 27. Neo4j, Inc. All rights reserved 2022 27 Neo4j is Part of your Data Ecosystem DATA SOURCES USE CASES INGEST Apache Hop Structured Unstructured DATA ANALYTICS DATA MANAGEMENT Journey Analytics Risk Analytics Churn Analysis What-if Analysis Feature Engineering & ML Fraud Recommendations Data Fabric Data Compliance Data Governance Data Provenance Data Lineage Next Best Case Ontologies Neo4j Bloom Neo4j GDS Library PRODUCT COMPONENTS APOC VISUALIZE AUTO ML DRIVERS & APIs
- 28. Neo4j, Inc. All rights reserved 2022 © 2022 Neo4j, Inc. All rights reserved. 28 Thank you!