About this document

CUBOS: Cluster Validity for Categorical Data by orca is a document available to read on EtoBox.

This document proposes a new internal cluster validity index called CUBOS for evaluating clustering performance on categorical data. CUBOS improves upon existing indices by using a new distance metric called IDC that considers relationships between categorical attribute values. IDC allows CUBOS to measure distances between categorical data objects directly and explore more detailed distribution information in clustering results. The experimental results on UCI datasets show CUBOS can produce more accurate e

Author

orca

Language

EN