Vector Databases
Data systems optimized for storing embeddings and retrieving semantically similar text, images, audio, code, and other high-dimensional data.
Core metadata
- ID: vector_databases
- Era: Modern
- First known date: 2019 (decade)
- Region: Global software industry
- Review status: source_checked
- Maturity: established
Prerequisites
- Databases (Relational DBMS) (databases_relational_dbms)
- Information Theory (information_theory)
- Machine Learning (Early Algorithms) (machine_learning_early_algorithms)
Dependents
- None.
Fields
Field lanes
- Artificial Intelligence & Machine Learning: Deployment & MLOps
Node sources
- Lucene for Approximate Nearest-Neighbors Search on Arbitrary Dense Vectors (arXiv, 2019, primary_paper) • Supports: node, edge, maturity
Prerequisite edge evidence
Edge/source evidence summary:
- Prerequisite edges: 3
- Average edge confidence: 70%
- Prerequisite sources: 3
- expert_inference: 3
| Prerequisite | Type | Confidence | Evidence level | Note | Sources |
|---|---|---|---|---|---|
| Databases (Relational DBMS) (databases_relational_dbms) | enabling | 68% | expert_inference | Databases (Relational DBMS) provides a capability that enables this technology without being the only possible path. |
|
| Machine Learning (Early Algorithms) (machine_learning_early_algorithms) | enabling | 75% | expert_inference | Embedding models and nearest-neighbor methods make vector databases useful, but LLMs are not required. |
|
| Information Theory (information_theory) | enabling | 68% | expert_inference | Information Theory provides a capability that enables this technology without being the only possible path. |
|
This page is generated from canonical era JSON and is indexable by URL.