This Week in Neo4j: Graph Embeddings, OSS Vulnerabilities, Supply Chain Stack, Star Wars Graph, Chatbot MD, and More!

With graph data science, you can choose between several graph embedding algorithms, and generating low-dimensional vector spaces is easy. But how do you evaluate the embedding results in your machine learning application? Enter Emblaze, a Jupyter notebook widget for visually comparing embeddings developed by Carnegie Mellon University’s Data Interaction Group. In his tutorial, Nathan Smith shows you the power of the widget’s Python API to perform dimensionality reduction on multiple sets of embedding data and compare the embeddings using animated scatter plots.

Don’t miss the tutorial: Understanding Graph Embeddings with Neo4j and Emblaze

Yolande Poirier

Estelle is CTO & co-founder of SmartGrid. She started out in nuclear and particle physics working at CERN before applying her data science skills to geo-spatial property and transportation projects. As a heavy user of graph databases, she focuses on building efficient machine learning models. She recently authored the book Hands-On Graph Analytics with Neo4j. Kudos to Estelle! Follow her on Twitter.
GRAPH EMBEDDINGS: Understanding Graph Embeddings With Neo4j and Emblaze
Say you’ve generated graph embedding vectors for your machine learning pipeline. How do you find the best one for your application? In this tutorial with Nathan Smith, you will generate multiple graph embeddings and then use Emblaze, a Jupyter notebook widget, to visualize and compare the results.
SECURITY: How Neo4j’s Graph Database Can Remediate Vulnerabilities
See how the team at Debricked is using graph technology to map the deep dependencies that package managers miss. They find safe versions of root dependencies of open source software, automating the update process.
VIDEO: Exploring a Star Wars Dataset in Neo4j

In this step-by-step demo, Jonathan Thein shows you how to go from importing to exploring data within seconds — all without writing a single line of code.

BLOG: Data Stack at Scoutbee – Episode 1

This article offers a fun and fascinating inside view of how a big data company in the supply chain industry is building their data and ML stack with AWS, Neo4j Aura, and Databricks.

CHATBOT: Navigate a Medical Knowledge Graph Using English
Sixing Huang presents, a multilingual voice chatbot that is using a knowledge graph to get medical questions answered in the language of your choice – without a doctor!
TUTORIAL: Getting Started With a Graph Database
Ved Vasu Sharma helps you set up a Neo4j Graph Database on your local system and start running queries with Cypher.
TWEET OF THE WEEK: @javiercha
Don’t forget to retweet if you like it!

Thomas Larsen released an update to his Neo4j connectors for Alteryx. You can now output multiple relationships between nodes of the same type. Learn more.

Neo4j/GraphQL and Apollo Federation repo demonstrates adding bare-bones Apollo Federation subgraph specs to a Neo4j/graphql app. Check out it on the GitHub .

Yusuf Canbaz wrote his Master’s thesis about generic and customizable graph visualization. It also includes some graph traversal algorithms. Learn more.