Vector similarity search is one of the fastest-growing domains in AI and machine learning. At its core, it is the process of matching relevant pieces of information together.

Similarity search is a complex topic and there are countless techniques for building effective search engines.

In this video, we'll cover three vector-based approaches for comparing languages and identifying similar 'documents', covering both vector similarity search and semantic search:

- TF-IDF
- BM25
- Sentence-BERT

📰 Original article:
[ Ссылка ]

🤖 70% Discount on the NLP With Transformers in Python course:
[ Ссылка ]

🎉 Sign-up For New Articles Every Week on Medium!
[ Ссылка ]

Mining Massive Datasets Book (Similarity Search):
📚 [ Ссылка ] (3rd ed)
📚 [ Ссылка ] (1st ed, cheaper)

👾 Discord
[ Ссылка ]

🕹️ Free AI-Powered Code Refactoring with Sourcery:
[ Ссылка ]

00:00 Intro
01:37 TF-IDF
11:44 BM25
20:30 SBERT

свернуть