Better Semantic Search with Hybrid Sparse-Dense Search - Haystack US 2023
Offered By: OpenSource Connections via YouTube
Course Description
Overview
Explore the advantages of hybrid search in this conference talk from Haystack US 2023. Delve into the world of semantic search as Roie Schwaber Cohen, a staff developer advocate at Pinecone, discusses the combination of sparse and dense embedding models. Learn about the limitations of traditional keyword-based algorithms like TF-IDF and BM25, and discover how dense embedding models enable more natural querying based on semantic meaning. Understand the challenges of using dense models in data-scarce environments with domain-specific terminology. Gain insights into how hybrid search leverages the strengths of both sparse and dense embeddings to enhance vector search accuracy and applicability across various scenarios. Discover practical approaches to building improved semantic search applications using hybrid search techniques.
Syllabus
Haystack US 2023 - Roie Schwaber Cohen: Better Semantic Search with Hybrid (Sparse-Dense) Search
Taught by
OpenSource Connections
Related Courses
U&P AI - Natural Language Processing (NLP) with PythonUdemy What's New in Cognitive Search and Cool Frameworks with PyTorch - Episode 5
Microsoft via YouTube Stress Testing Qdrant - Semantic Search with 90,000 Vectors - Lightning Fast Search Microservice
David Shapiro ~ AI via YouTube Semantic Search for AI - Testing Out Qdrant Neural Search
David Shapiro ~ AI via YouTube Spotify's Podcast Search Explained
James Briggs via YouTube