Ranking inter-relationships between clusters

• Published in: International journal of systems science Vol. 42; no. 12; pp. 2071 - 2083
• Main Authors: Wang, Tingting, Chen, Feng, Chen, Yi-Ping Phoebe
• Format: Journal Article
• Language: English
• Published: Abingdon Taylor & Francis Group 01.12.2011; Taylor & Francis
• Subjects: Algorithms; Applied sciences; Artificial intelligence; Biological and medical sciences; Biology; Cancer; Classification; Clusters; Computer science; control theory; systems; conceptual clusters; Criteria; Data processing. List processing. Character string processing; Evolution; Exact sciences and technology; Information theory; Information, signal and communications theory; Medical sciences; Memory organisation. Data processing; mode pattern; mutual information; Pattern recognition. Digital image processing. Computational geometry; relationship tree; Software; Telecommunications and information theory; Tumors; Human; Tree(graph); Hierarchical classification; Image databank; mode pattern; Cluster; Metallurgical industry; Biology; Malignant tumor; Data mining; conceptual clusters; Information measure; Cancerology; relationship tree; Diagnosis; Mutual information; Cancer
• ISSN: 0020-7721; 1464-5319
• DOI: 10.1080/00207721003710649

The evaluation of the relationships between clusters is important to identify vital unknown information in many real-life applications, such as in the fields of crime detection, evolution trees, metallurgical industry and biology engraftment. This article proposes a method called 'mode pattern + mutual information' to rank the inter-relationship between clusters. The idea of the mode pattern is used to find outstanding objects from each cluster, and the mutual information criterion measures the close proximity of a pair of clusters. Our approach is different from the conventional algorithms of classifying and clustering, because our focus is not to classify objects into different clusters, but instead, we aim to rank the inter-relationship between clusters when the clusters are given. We conducted experiments on a wide range of real-life datasets, including image data and cancer diagnosis data. The experimental results show that our algorithm is effective and promising.