Building Recommendation Systems with Vector Search - Women of Search
OpenSource Connections via YouTube
Overview
Explore the intersection of search technology and recommendation systems in this insightful conference talk from Haystack US 2023. Gain valuable insights into the Women of Search initiative and learn how to build effective recommendation systems using vector search. Discover the significant impact of poor search and recommendation on e-commerce businesses, and understand the importance of proper item and user representation. Delve into Ref2Vec, a feature in Weaviate that converts users to vectors, and learn how to construct bipartite graphs for creating digital fingerprints. Understand the process of achieving recommendations through user vector queries and gain practical knowledge on implementing these techniques. Benefit from Erika Cardenas' expertise in vector databases, data science, and economics as she shares her experience and research. Engage in a Q&A session covering both the presentation content and the Women of Search initiative, providing an opportunity to deepen your understanding of these crucial topics in the field of search technology.
Syllabus
Introduction
How to join
About Erika
Agenda
What is Recommendation
User Preferences
ContentBased Filtering
Collaborative Filtering
Limitations
Vector Representations
Recommendation Representations
Demo
Demo Repository
Discussion Topics
Questions
Wrapup
Taught by
OpenSource Connections