Overview

In Casibase, vectors are one of its core strengths. Vector technology plays a key role in knowledge representation and retrieval, and by pairing it with the stores feature, which converts data such as text and images into dense vectors, Casibase enables efficient similarity search and data analysis.

For information on the definition of vectors, see the core-concepts section in our previous documentation.

Application of vector technology in Casibase

Knowledge Embedding

Users can upload files in various formats (e.g. TXT, Markdown, Docx, PDF, etc.) and select embedding methods (e.g. Word2Vec, GloVe, BERT, etc.) to generate knowledge and corresponding vectors. These vectors are stored in a vector database for quick retrieval and query.

Similarity Search

Casibase converts the knowledge into vectors and stores them in a vector database. This vector representation supports a powerful similarity search function, which allows users to quickly find relevant information based on context or content.