Embedding clustering

Embedding clustering groups documents, queries, or users by embedding similarity — used for topic discovery, deduplication, semantic indexing, and personalisation.

Once you have embeddings for a corpus, clustering them (k-means, HDBSCAN, agglomerative) reveals natural topic structure without supervision. Common applications in 2026 include: discovering FAQ clusters in support tickets, deduplicating near-identical documents before ingestion, building topic taxonomies for content sites, segmenting users for personalisation, and routing queries to specialised models or agents. Combined with dimensionality reduction (UMAP, t-SNE) for visualisation, embedding clustering is the standard exploratory technique for any embedding-based system.

When to use embedding clustering

• Topic discovery in unsupervised corpora.
• Document deduplication.
• User / query segmentation for routing.

Common mistakes

• Clustering without normalising the embeddings — distance metrics misbehave.
• Using k-means when HDBSCAN would fit the topic shape better.

FAQ

Related terms

• Embeddings — Embeddings are dense numeric vectors that represent the meaning of text, images, or other data, allowing similarity to be measured as vector distance.
• Semantic search — Semantic search finds documents by meaning rather than keyword match, using embedding similarity in a vector space.
• Semantic routing — Semantic routing classifies an incoming query by meaning — via embedding similarity to predefined route prototypes — and dispatches it to the right model, agent, or sub-system.

Last updated: 2026-06-01. Raw markdown: https://promtable.com/glossary/embedding-clustering.md.