Similarity for Binary Data Based on the Value of Entropy and Structure Patterns Categories

Abstract

Similarity of two objects that have a form of binary data, usually calculated based on the frequencies in thecontingency table that includes all discrete random variables. In this article we will discuss the similaritymeasures for binary data based on entropy values and structural patterns of the two object categories. Measuringsimilarity based on the value of entropy and structural pattern of categories can be used as a validation measureof similarity for binary data.