Unsupervised Clustering - AFS

AFS Installation and Administration Guide

Product
AFS
AFS_Version
7.10
Platform
RHEL
Category
Reference Guide

Unsupervised clustering consists in clustering results in result groups based on the document similarity calculated through vectorization.

It means that documents are clustered based on how similar they are. Similarity is calculated by comparing document vectors, which are lists of prominent words in the document.

Unsupervised clustering is a kind of automatically-generated taxonomy.