This Week in Neo4j: Vector Index, Embeddings, Geospatial, Document QA and more

Welcome to This Week in Neo4j, your weekly fix for news from the world of graph databases!
This week, we take a hands-on approach to Neo4j Vector Search and LangChain, use Embeddings for improved Image Retrieval, look at some resources for GIS Plugin Building and watch a tutorial on Document QA with Neo4j.

Join our Neo4j Research panel! Sign up to share your experiences with a researcher and influence the future of Neo4j products.
What’s in it for you? A chance to connect directly with product development teams, get paid compensation, hear about what we are working on, and more!

I hope you enjoy this issue,
Alexander Erdl


Sharmistha Chatterjee is a Data Science Evangelist with vast professional experience in the field of Machine Learning and Cloud applications. She is passionate about technology, learning, and trying out anything new. In her work, she tries to bridge the gap between theory and practice when doing AI research and implementing scalable AI solutions for enterprises.
Connect with her on LinkedIn.

Sharmistha’s session at NODES “Graph Algorithms for Privacy and Fairness in the Healthcare Industry” brought forth how Responsible AI concepts can be extended to graph algorithms in the healthcare industry. She highlighted two important pillars of Responsible AI, privacy and fairness, which help deal with fraud and bias in the healthcare field.

Sharmistha Chatterjee

VECTOR INDEX: Neo4j x LangChain: Deep dive into the new Vector index implementation

Neo4j was and is an excellent fit for handling structured information and got even better with the introduction of a new vector index in version 5.11 designed to efficiently perform semantic search over unstructured text or other embedded data modalities. This blog post by Tomaz Bratanic is designed to walk you through all the customization options available in the Neo4j Vector Index implementation in LangChain.
EMBEDDIINGS: Revolutionizing Image Retrieval: Harnessing Neo4j and Embeddings for Speed and Precision
Image retrieval discovers images resembling a given query image within a massive dataset. In the past, this was often done by comparing pixel values directly – a slow and not-so-accurate process. But fear not; there’s a smarter way.We’re talking about embeddings and cosine similarity – the dynamic duo of modern image retrieval. Embeddings are like magic codes representing images in a compact, feature-packed format. This post by Wickkiey gives you a rundown how it can be done.
GEOSPATIAL: Building Open Source GIS Plugins
Will Lyon shares his slides from a talk he held at FOSS4G America a few weeks ago. He covers why to build QGIS Plugins, including an in-depth explanation of the QGIS Ecosystem and links to handy tutorials on how to build your own QGIS Plugin.

VIDEO: Implementing Document QA Neo4j & VertexAI: Step By Step Introduction To get LLM To Work With Neo4j

This video discusses how to get access to VertexAI API Endpoints for Embedding & Completion PALM Model, covers all the necessary steps to get your Neo4j database ready with the necessary plugins etc to embed the text sentence inside Neo4j DB and then store it for later retrieval. Kamalraj then uses the Graph Data Science module to execute a similarity search on the stored Embedding and user query

TWEET OF THE WEEK: Othmane Zoheir

Don’t forget to share it if you like it!