Long-Term Knowledge Storage and Retrieval › Glossary

Embedding: A numerical representation (vector) of text that captures its semantic meaning. Similar meanings produce similar vectors, enabling semantic search.

Key-Value Store: A simple storage system where data is saved with a label (key) and retrieved using that same label. Like a dictionary or hash map.

Knowledge Base: A structured collection of information that an agent can search and retrieve from. More sophisticated than simple key-value storage.

Semantic Search: Finding information based on meaning rather than exact keyword matches. Uses embeddings to understand that "What should I avoid eating?" relates to allergy information.

Vector Database: A specialized database optimized for storing and searching embeddings. It supports fast similarity search across large collections of data.

Vector Store: Another term for vector database. A system that stores embeddings and supports similarity-based retrieval.

Cosine Similarity: A mathematical measure of how similar two vectors are, ranging from 0 (completely different) to 1 (identical). Used to find relevant information in vector search.