To search, Click below search items.


All Published Papers Search Service


Consideration of Site-wise Confidence in Fuzzy Co-clustering of Vertically Distributed Cooccurrence Data


Toshiya Oda, Katsuhiro Honda, Seiki Ubukata, Akira Notsu


Vol. 16  No. 2  pp. 15-22


In advanced information and telecommunications network society, it is expected to utilize big data distributed among various organizations, such as cooperation groups, state organs and allied countries, with the goal of revealing intrinsic knowledge. In such collaborative data mining, however, personal privacy must be strictly preserved. This paper deals with a possible approach for utilizing distributed cooccurrence information in fuzzy co-clustering context under privacy consideration. Fuzzy Clustering for Categorical Multivariate data (FCCM) is a basic fuzzy co-clustering model and have been extended so as to perform privacy preserving data analysis. The secure model is further improved in this paper so that we can find robust knowledge, which is free from the influences of unreliable site, considering site-wise confidences. The applicability of the proposed model is demonstrated in several numerical experiments.


Fuzzy Clustering, Co-clustering, Privacy preserving data mining.