期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2021
卷号：99
期号：11
语种：English
出版社：Journal of Theoretical and Applied
摘要：Clustering still leaves problems in selecting optimal clusters in order to obtain a right and correct classification analysis. Right in the sense of the number of clusters, while correct in terms of the information generated by a group of cluster members that is optimally grouped. Determining the optimal number of clusters is a difficult problem in non-polynomials. A number of existing approaches generally still rely on the number of K tests tested. This study aims to produce a new approach that can determine and place data in clusters optimally in a combinatorial form. This can be done by considering that the problem of selecting cluster placement has a combinatorial optimization structure pattern. However, the resulting combinatorial optimization model is quadratic. Therefore, in order to make the combinatorial clustering problem easier to solve, linearization of the cluster data was carried out so that a combinatorial optimization approach was produced with the algorithm. Several illustrations have been put forward to demonstrate the validity of the method. The combinatorial optimization approach as proposed in this research produces novelty on cluster data analysis techniques.
关键词：Clustering;Information;Combinatorial Optimization;Linearization;Cluster Da
