What is graph all about, and why should you care? Graphs come in many shapes and forms, and can be used for different applications: Graph Analytics, Graph AI, Knowledge Graphs, and Graph Databases. Talk by George Anadiotis. Connected Data London Meetup June 29th 2020. Up until the beginning of the 2010s, the world was mostly running on spreadsheets and relational databases. To a large extent, it still does. But the NoSQL wave of databases has largely succeeded in instilling the “best tool for the job” mindset. After relational, key-value, document, and columnar, the latest link in this evolutionary proliferation of data structures is graph. Graph analytics, Graph AI, Knowledge Graphs and Graph Databases have been making waves, included in hype cycles for the last couple of years. The Year of the Graph marked the beginning of it all before the Gartners of the world got in the game. The Year of the Graph is a term coined to convey the fact that the time has come for this technology to flourish. The eponymous article that set the tone was published in January 2018 on ZDNet by domain expert George Anadiotis. George has been working with, and keeping an eye on, all things Graph since the early 2000s. He was one of the first to note the continuing rise of Graph Databases, and to bring this technology in front of a mainstream audience. The Year of the Graph has been going strong since 2018. In August 2018, Gartner started including Graph in its hype cycles. Ever since, Graph has been riding the upward slope of the Hype Cycle. The need for knowledge on these technologies is constantly growing. To respond to that need, the Year of the Graph newsletter was released in April 2018. In addition, a constant flow of graph-related news and resources is being shared on social media. To help people make educated choices, the Year of the Graph Database Report was released. The report has been hailed as the most comprehensive of its kind in the market, consistently helping people choose the most appropriate solution for their use case since 2018. The report, articles, news stream, and the newsletter have been reaching thousands of people, helping them understand and navigate this landscape. We’ll talk about the Year of the Graph, the different shapes, forms, and applications for graphs, the latest news and trends, and wrap up with an ask me anything session.

1. THEYEARS OFTHE GRAPH:
THE FUTURE OFTHE FUTURE IS HERE
George Anadiotis
Connected Data London Meetup, June 29th 2020

2. ABOUT ME
 Working with data since 1992
 Graph since early 2000
 Databases
 Modeling
 Research
 Analysis
 Consulting
 Entrepreneurship
 Journalism

3. THEYEAR OFTHE GRAPH:
THE GO-TO SOURCE FOR ALLTHINGS GRAPH
 Term and article
 * Published on ZDNet in January 2018
 * Before the hype
 Site
 * https://yearofthegraph.xyz/
 Newsletter
 * https://yearofthegraph.xyz/newsletter/
 Social Media
 * https://www.linkedin.com/showcase/43364427/
 * https://twitter.com/linked_do
 Graph Database Report
 * https://yearofthegraph.xyz/graph-database-report/

4. WHAT IS GRAPH?
Graph Analytics |Knowledge Graphs | Graph DBs | Graph AI

5. GRAPH ANALYTICS:
FROMTHE BRIDGES OF KÖNIGSBERGTO
MODERN DATA SCIENCE

6. GRAPH ANALYTICS:
PATHFINDING AND GRAPH SEARCH ALGORITHMS
 Search
 * Explore a graph either for general
discovery or explicit search
 * Example: Locate neighbors
 Pathfinding
 * Explore routes between nodes
 * Example: Navigation
 Graph Algorithms: Practical Examples in Apache Spark
and Neo4j. Mark Needham, Amy E. Hodler. O'Reilly 2019

7. GRAPH ANALYTICS:
CENTRALITY ALGORITHMS
 Centrality
 * Understand the roles of particular
nodes in a graph and their impact
on that network
 * Example: Find influence
 Graph Algorithms: Practical Examples in Apache Spark
and Neo4j. Mark Needham, Amy E. Hodler. O'Reilly 2019

8. GRAPH ANALYTICS:
COMMUNITY DETECTION ALGORITHMS
 Community Detection
 * Identifying related sets to reveal
clusters of nodes, isolated groups,
and network structure.
 * Example: Fraud analysis
 Graph Algorithms: Practical Examples in Apache Spark
and Neo4j. Mark Needham, Amy E. Hodler. O'Reilly 2019

9. GRAPH ANALYTICS: USE CASE
 Drug Discovery
 * Leading Pharma
 * Data on genes, proteins, etc
 * Identification of causal relationships

10. KNOWLEDGE GRAPHS:
FROMTIM BERNERS LEETO GOOGLE AND BEYOND

11. KNOWLEDGE GRAPHS:
GOOGLE, MEETTHE SEMANTICWEB
 From the Semantic Web to the world
 *The Web is a Graph, and Google based
its success on PageRank
 * Categorizing web content needs
metadata and semantics
 * Google adopted Semantic Web
technology, coined the term
Knowledge Graph
 * Besides Google’s Knowledge Graph,
everyone can have one
 * From Morgan Stanley to Average Jo
 * Personal Knowledge Graphs

12. KNOWLEDGE GRAPHS:
IT’S ALL ABOUT SEMANTICS AND SCHEMA

13. KNOWLEDGE GRAPHS:
KNOWLEDGE GRAPH = ONTOLOGY = AI
 Mark Hall, Executive Director at
Morgan Stanley
 *Traditional data modeling has
concerned itself primarily with the
capture and retrieval of data
 * Ontology concerns itself with a shared
understanding of what that data means
 * Before embarking on the AI-journey,
it’s critical to ensure you understand
and document your domain

14. KNOWLEDGE GRAPHS: USE CASE
 Knowledge Graph for Search
 * Leading Retailer in DACH
 * 200Million+ MAU, 300K+ search requests
 * Improve coverage, response time, bottom-line

15. GRAPH DATABASES:
LEVERAGING CONNECTIONS

16. GRAPH DATABASES:
MINDTHE HYPE
 The Practitioner's Guide to Graph Data: Applying Graph Thinking and Graph Technologies
to Solve Complex Problems. Denise Gosnell, Matthias Broecheler. O'Reilly 2019

17. GRAPH DATABASES:
WHAT ARETHEY? HOW DOYOU CHOOSE ONE?
 Operational vs. Analytical
 * Fully-fledged graph API
 * Operations & Analytics
 * Future-proof, integrated
 Native vs. Non-native
 *Designed as a graph database
 * Storing data in a native format
 * Optimized for graph

18. PROPERTY GRAPH DATABASE USE CASES
 Operational
 applications
 Graph
 Analytics
 AI

19. RDF GRAPH DATABASE USE CASES
 Data Integration  Knowledge Graph  AI

20. GRAPH DATABASE: USE CASE
 Smart Home - IoT
 * LeadingTelco in the Nordics
 * 1,5 Million Homes
 * Real-time processing

21. GRAPH AI:
THE FUTURE OFTHE FUTURE

22. GRAPH AI:
MACHINE LEARNING
 Image: Oracle

23. GRAPH AI:
GRAPH NEURAL NETWORKS
 Graph Neural Networks: A Review of Methods and Applications.
Zhou et. Al.
 Graph Neural Networks (GNNs)
 * Models that capture dependence
of graphs via message passing
between the nodes of graphs .
 * Unlike standard neural networks,
GNNs retain a state that can
represent information from its
neighborhood with arbitrary depth.
 * Domain knowledge can effectively
help a deep learning system
bootstrap its knowledge, by
encoding primitives instead of
forcing the model to learn these
from scratch.

24. GRAPH AI:
GRAPH EMBEDDINGS
 Image: Oracle
 Graph Embeddings
 * Embeddings: reduce dimensions of
input to machine learning algorithms
 * Graph type data are discrete. Graph
embedding pre-processes graphs to
turn them into a continuous vector
space.
 * Walk embedding methods perform
graph traversals with the goal of
preserving structure and features
 * Proximity embedding methods use
Deep Learning methods and/or
proximity loss functions to optimize
proximity

25. GRAPH AI: USE CASE
 Anti-Fraud in real-time
 * LeadingTelco in China
 * 600 Million Users
 * Compliance, trust

26. GRAPHS ARE EVERYWHERE

27. THEYEAR OFTHE GRAPH:
THE GO-TO SOURCE FOR ALLTHINGS GRAPH
 Term and article
 * Published on ZDNet in January 2018
 * Before the hype
 Site
 * https://yearofthegraph.xyz/
 Newsletter
 * https://yearofthegraph.xyz/newsletter/
 Social Media
 * https://www.linkedin.com/showcase/43364427/
 * https://twitter.com/linked_do
 Graph Database Report
 * https://yearofthegraph.xyz/graph-database-report/