A New Context-Based Similarity Measure for Categorical Data Using Information Theory.
Thanh-Phu NguyenMina RyokeVan-Nam HuynhPublished in: IUKM (2018)
Keyphrases
- categorical data
- information theory
- similarity measure
- information theoretic
- mutual information
- hierarchical clustering algorithm
- cluster analysis
- numerical data
- parameter free
- jensen shannon divergence
- clustering method
- statistical learning
- numeric data
- attribute values
- statistical mechanics
- distance measure
- normalized mutual information
- pairwise
- conditional entropy
- distance based outlier detection
- shannon entropy
- real world
- mdl principle
- image processing