A Real-World Guide to Building
Your Knowledge Graphs
Nav Mathur
Sr. Director – Global Solutions, Neo4j, Inc.
in/navmathur
@nav_mathur
2
What do these organizations have in Common?
2
3
More real-world knowledge graphs
4
Knowledge Graph Vs Knowledge Base
“Unlike a simple knowledge base with flat structures and static
content, a knowledge graph acquires and integrates adjacent
information using data relationships to derive new knowledge.”
Connected around
relevant attributes.
(contextually
related)
Dynamically
updating / not
manual
Uses intelligent
labelling and ties in
to the graph
automatically
Explainable -
Intelligent metadata
helps traverse to
find answers to
specific problems,
even when we don’t
know exactly how
to ask for it.
Usually contains
heterogeneous data
types. It combines
and uncovers
connections across
silos of information.
Key Principles of a Knowledge Graph
6
Financial Services Knowledge Graph
Credit Risk Management
7
Financial Services Knowledge Graph
Investment Risk Management
8
Financial Services Knowledge Graph
Portfolio News Recommendations
9
Portfolio News Recommendations
Data Model
User
Org
Article
Topic
Topic
Group
HAS_TOPIC
GROUP
WEIGHT
TRACKS
REFERENCES
IS_AFFILIATED
SUPPLIES
HAS_ULTIMATE_OWNER
HAS_IMMIDIATE_OWNER
10
Data Orchestration Layer
Data Sources
CLIENT Admin Dashboard
Session
Data
Feedback
Scored
Recommen-
dations
Graph
Algorithms
AI / ML
Click
Stream
Data
INTELLIGENT RECOMMENDATIONS FRAMEWORK
Discovery
Exclude
Boost
Diversity
User Segmentation
Item Similarity
Recommendation Engines
Building your KG
• Modelling
• Data Ingestion
• Auto Labelling (NLP)
• Scoring
• Data Lineage
• Alerting
• Auto and human aided
merging/similarity
• Integrate ML for
refreshing /updating
the graph
RSS Feed
Org. Feed
(Graph)
Portfolio News
Recommendations
Demo
12
Recommendations Are Everywhere
Job Search
and Recruiting
Financial
Services
Retailing
Government
Services
Healthcare
Travel
Media and
Entertainment
13
Recommendations Are Everywhere in the Enterprise
Human Resources
Supplier Analytics
Product
and BOM
Analytics
Personalization
Customer
Journey
14
Neo4j HCM Use Cases
Ratings
Normalization
Succession
Planning
Building Cross-
Functional TeamsFlight Risk
Lifetime
Employee Value
Promotion and
Compensation
Recommendations
Retail Recommendations
Filter Criteria
Price, brand, color…
Similar Products
Automated bundling
and pricing
Related Products
Customer 360 / Customer Journey Recommendations
Profitability and Margin
Analysis
Dynamic
CSAT Score
Customer Acquisition
and Retention Cost
Lifetime
Customer Value
Engagement
Analysis and Alerts
Churn Score
and Analysis
Personal 360 Recommendations
Building
Communities / BOFs
Engagement
Score
People
Connections
CCPA/GDPR
Content
Recommendations
Discover
cohorts
Customers
Product and BOM Recommendations
Bottleneck
analysis
Optimized plant
assignment
MBOM
analysis
Part/material
similarity
Inventory
analysis
Alternate vendors
and suppliers
Influential part /
material discovery
19
Hybrid Scoring-Based Approach is More Contextual
Graph technology enables you to make
recommendations that weight multiple methods
Collaborative
Filtering
Based on similar
users or products
Content
Filtering
Based on user
history and profile
Rules-Based
Filtering
Based on
predefined rules
and criteria
Business
Strategy
Based on
promotions,
margins, inventory
20
Neo4j Gives You Control of Your Online Business
Self-learning framework improves recommendations over time
Customer Context
Recommendations
Monitor and Adjust
Machine Learning
Feedback
Graph Algorithms
Pathfinding and Search
Parallel Breadth
First Search and DFS
Shortest Path
Single-Source Shortest Path
All Pairs Shortest Path
Minimum Spanning Tree
A* Shortest Path
Yen’s K Shortest Path
K-Spanning Tree MST
Centrality and Importance
Degree Centrality
Closeness Centrality
Betweenness Centrality
PageRank
Harmonic Closeness Centrality
Dangalchev Closeness Centrality
Wasserman and Faust Closeness
Centrality
Approximate Betweenness Centrality
Personalized PageRank
Community Detection
Triangle Count
Clustering Coefficients
Connected Components
Union Find
Strongly Connected
Components
Label Propagation
Louvain Modularity
1 Step
Balanced Triad
Identification
Louvain
Multi-Step
Similarity and
ML Workflow
Euclidean Distance
Cosine Similarity
Jaccard Similarity
Random Walk
One Hot Encoding
22
Data Sources
CLIENT Admin Dashboard
Session
Data
Feedback
Scored
Recommen-
dations
Graph
Algorithms
AI / ML
Click
Stream
Data
INTELLIGENT RECOMMENDATIONS FRAMEWORK
Discovery
Exclude
Boost
Diversity
User Segmentation
Item Similarity
Intelligent Recommendations Framework
Recommendation Engines
Recommendation Framework Technology
Engines Are Processing Pipelines
• Pipelines mimic process of building
recommendations to reduce query complexity
• Easier to develop and maintain
• Can enable and disable different parts of the
pipeline based on business rules
Promotions, inventory, etc.
• Ability to score and weight phases differently
• Supports traceability and explainability
Recommendation
Framework Advantages
• Faster time to realization
• Development and
implementation flexibility
• Code-free development
Highly Configurable Engines
• Works with any data model
• Highly contextual recommendations
for what the user is doing now
• Different engines for different uses
Discovery
Exclude
Boost
Diversity
Thank You
• Nav Mathur
• Sr. Director – Global Solutions, Neo4j, Inc.
• in/navmathur
• @nav_mathur
25
720+
7/10
12/25
8/10
53K+
100+
300+
450+
Adoption
Top Retail Firms
Top Financial Firms
Top Software Vendors
Customers Partners
• Creator of the Neo4j Graph Platform
• ~250 employees
• HQ in Silicon Valley, other offices include
London, Munich, Paris and Malmö Sweden
• $80M new funding led by Morgan Stanley & One
Peak. Total $160M from Fidelity, Sunstone,
Conor, Creandum, and Greenbridge Capital
• Over 15M+ downloads & container pulls
• 300+ enterprise subscription customers
with over half with >$1B in revenue
Ecosystem
Startups in program
Enterprise customers
Partners
Meet up members
Events per year
Industry’s Largest Dedicated Investment in Graphs
Neo4j - The Graph Company

Real World Guide to Building Your Knowledge Graph

  • 1.
    A Real-World Guideto Building Your Knowledge Graphs Nav Mathur Sr. Director – Global Solutions, Neo4j, Inc. in/navmathur @nav_mathur
  • 2.
    2 What do theseorganizations have in Common? 2
  • 3.
  • 4.
    4 Knowledge Graph VsKnowledge Base “Unlike a simple knowledge base with flat structures and static content, a knowledge graph acquires and integrates adjacent information using data relationships to derive new knowledge.”
  • 5.
    Connected around relevant attributes. (contextually related) Dynamically updating/ not manual Uses intelligent labelling and ties in to the graph automatically Explainable - Intelligent metadata helps traverse to find answers to specific problems, even when we don’t know exactly how to ask for it. Usually contains heterogeneous data types. It combines and uncovers connections across silos of information. Key Principles of a Knowledge Graph
  • 6.
    6 Financial Services KnowledgeGraph Credit Risk Management
  • 7.
    7 Financial Services KnowledgeGraph Investment Risk Management
  • 8.
    8 Financial Services KnowledgeGraph Portfolio News Recommendations
  • 9.
    9 Portfolio News Recommendations DataModel User Org Article Topic Topic Group HAS_TOPIC GROUP WEIGHT TRACKS REFERENCES IS_AFFILIATED SUPPLIES HAS_ULTIMATE_OWNER HAS_IMMIDIATE_OWNER
  • 10.
    10 Data Orchestration Layer DataSources CLIENT Admin Dashboard Session Data Feedback Scored Recommen- dations Graph Algorithms AI / ML Click Stream Data INTELLIGENT RECOMMENDATIONS FRAMEWORK Discovery Exclude Boost Diversity User Segmentation Item Similarity Recommendation Engines Building your KG • Modelling • Data Ingestion • Auto Labelling (NLP) • Scoring • Data Lineage • Alerting • Auto and human aided merging/similarity • Integrate ML for refreshing /updating the graph RSS Feed Org. Feed (Graph)
  • 11.
  • 12.
    12 Recommendations Are Everywhere JobSearch and Recruiting Financial Services Retailing Government Services Healthcare Travel Media and Entertainment
  • 13.
    13 Recommendations Are Everywherein the Enterprise Human Resources Supplier Analytics Product and BOM Analytics Personalization Customer Journey
  • 14.
    14 Neo4j HCM UseCases Ratings Normalization Succession Planning Building Cross- Functional TeamsFlight Risk Lifetime Employee Value Promotion and Compensation Recommendations
  • 15.
    Retail Recommendations Filter Criteria Price,brand, color… Similar Products Automated bundling and pricing Related Products
  • 16.
    Customer 360 /Customer Journey Recommendations Profitability and Margin Analysis Dynamic CSAT Score Customer Acquisition and Retention Cost Lifetime Customer Value Engagement Analysis and Alerts Churn Score and Analysis
  • 17.
    Personal 360 Recommendations Building Communities/ BOFs Engagement Score People Connections CCPA/GDPR Content Recommendations Discover cohorts
  • 18.
    Customers Product and BOMRecommendations Bottleneck analysis Optimized plant assignment MBOM analysis Part/material similarity Inventory analysis Alternate vendors and suppliers Influential part / material discovery
  • 19.
    19 Hybrid Scoring-Based Approachis More Contextual Graph technology enables you to make recommendations that weight multiple methods Collaborative Filtering Based on similar users or products Content Filtering Based on user history and profile Rules-Based Filtering Based on predefined rules and criteria Business Strategy Based on promotions, margins, inventory
  • 20.
    20 Neo4j Gives YouControl of Your Online Business Self-learning framework improves recommendations over time Customer Context Recommendations Monitor and Adjust Machine Learning Feedback
  • 21.
    Graph Algorithms Pathfinding andSearch Parallel Breadth First Search and DFS Shortest Path Single-Source Shortest Path All Pairs Shortest Path Minimum Spanning Tree A* Shortest Path Yen’s K Shortest Path K-Spanning Tree MST Centrality and Importance Degree Centrality Closeness Centrality Betweenness Centrality PageRank Harmonic Closeness Centrality Dangalchev Closeness Centrality Wasserman and Faust Closeness Centrality Approximate Betweenness Centrality Personalized PageRank Community Detection Triangle Count Clustering Coefficients Connected Components Union Find Strongly Connected Components Label Propagation Louvain Modularity 1 Step Balanced Triad Identification Louvain Multi-Step Similarity and ML Workflow Euclidean Distance Cosine Similarity Jaccard Similarity Random Walk One Hot Encoding
  • 22.
    22 Data Sources CLIENT AdminDashboard Session Data Feedback Scored Recommen- dations Graph Algorithms AI / ML Click Stream Data INTELLIGENT RECOMMENDATIONS FRAMEWORK Discovery Exclude Boost Diversity User Segmentation Item Similarity Intelligent Recommendations Framework Recommendation Engines
  • 23.
    Recommendation Framework Technology EnginesAre Processing Pipelines • Pipelines mimic process of building recommendations to reduce query complexity • Easier to develop and maintain • Can enable and disable different parts of the pipeline based on business rules Promotions, inventory, etc. • Ability to score and weight phases differently • Supports traceability and explainability Recommendation Framework Advantages • Faster time to realization • Development and implementation flexibility • Code-free development Highly Configurable Engines • Works with any data model • Highly contextual recommendations for what the user is doing now • Different engines for different uses Discovery Exclude Boost Diversity
  • 24.
    Thank You • NavMathur • Sr. Director – Global Solutions, Neo4j, Inc. • in/navmathur • @nav_mathur
  • 25.
    25 720+ 7/10 12/25 8/10 53K+ 100+ 300+ 450+ Adoption Top Retail Firms TopFinancial Firms Top Software Vendors Customers Partners • Creator of the Neo4j Graph Platform • ~250 employees • HQ in Silicon Valley, other offices include London, Munich, Paris and Malmö Sweden • $80M new funding led by Morgan Stanley & One Peak. Total $160M from Fidelity, Sunstone, Conor, Creandum, and Greenbridge Capital • Over 15M+ downloads & container pulls • 300+ enterprise subscription customers with over half with >$1B in revenue Ecosystem Startups in program Enterprise customers Partners Meet up members Events per year Industry’s Largest Dedicated Investment in Graphs Neo4j - The Graph Company