Clustering Data with Categorical Relationships

data aims at considering numeric data, categorical data or a mixture of both. For such data, the concentration was finding a relationship between the data points to be clustered. The relationships were limited to being either binary or fuzzy. Both involved a numeric value called distance or any other similarity measure between two data points and cluster them together if found similar. With time, a new kind of relationship called categorical relationship was observed between data points, far different from the traditionally seen ones. This paper focuses on handling data points having categorical relationships and the techniques emerged till date in this direction.