Rough K-prototypes Clustering Algorithm Based on OTC Similarity and Between-Cluster Information for Mixed Data

Abstract

In the iterative updating process of cluster center and partition matrix, most of the clustering algorithms for mixed data mainly consider the impact of the information within the cluster, but ignore the information between clusters, which results in low cluster separation. A rough clustering algorithm for mixed data based on the frequency division information between clusters is proposed. In the process of measuring the similarity between data objects and clusters, the unified object-cluster (OTC) similarity of mixed attribute data is used to avoid the transformation of classification attributes and numerical attributes and the parameter adjustment in the traditional clustering algorithms for mixed attribute data. The inter-cluster frequency division information is added in the iteration process of algorithm to ensure that the clustering result has high intra-cluster compactness and inter-cluster separation. Several groups of comparative experiments verified the effectiveness of the proposed algorithm.