Explore a comprehensive walkthrough of hybrid search implementation in Pinecone, combining semantic and keyword search capabilities. Learn how to leverage Pinecone's production-ready vector database for high-performance, scalable search solutions. Discover the step-by-step process of setting up hybrid search, including preprocessing, keyword creation, index creation, data upsert, and query setup. Dive into various search techniques such as basic keyword search, OR logic, AND logic, and negation. Gain practical insights into enhancing search functionality by integrating both semantic and traditional keyword search methods for more accurate and flexible results.
Overview
Syllabus
How Hybrid Search Works
Preprocessing
Creating Keywords
Creating an Index
Data Upsert
Query Setup
Keyword Search
OR Logic
AND Logic
Negation
Outro
Taught by
James Briggs
Related Courses
-
Vector Databases for Embeddings with Pinecone
-
Medical Search Engine with SPLADE + Sentence Transformers in Python
-
The Magic of Multilingual Search with Pinecone Serverless and Inference
-
Cohere AI's LLM for Semantic Search in Python
-
Vector and Hybrid Search with Elasticsearch
-
Supercharge eCommerce Search - OpenAI's CLIP, BM25, and Python
