Graph Machine Learning: Enhancing ML with Network Graphs and Embeddings

Explore graph machine learning techniques and their applications in this 59-minute conference talk from the Data Science Festival Summer School 2021. Discover how network graphs can reveal hidden relationships within data and enhance traditional machine learning models. Learn about graph analytics, graph-based machine learning, and the creation of graph embeddings to enrich existing ML pipelines. Follow along as Clair Sullivan, Data Science Advocate at Neo4j, demonstrates common ML problems and how to leverage graph-based approaches. Gain insights into graph modeling, database creation, and various graph algorithms including PageRank, node similarity, and community detection. Dive into practical examples using Jupyter Notebook and explore in-memory graph processing. Acquire valuable knowledge on cosine similarity and additional resources to further your understanding of graph machine learning.

Intro
Welcome
Overview
What is a graph
Why use a graph
How do you know you have a graph problem
Will it graph
Creating a graph
Graph modeling
Degrees betweenness page rank
Pathfinding example
Creating a database
Jupyter Notebook
Data Science
Page Rank
Write
Node Similarity
InMemory Graph
Community Detection
Character Detection
cosine similarity
other resources