MGR: یک الگوریتم خوشه بندی تفرقه سلسله مراتبی بر اساس تئوری اطلاعات برای داده های قطعی

Categorical data clustering has attracted much attention recently due to the fact that much of the data contained in today’s databases is categorical in nature. While many algorithms for clustering categorical data have been proposed, some have low clustering accuracy while others have high computational complexity. This research proposes mean gain ratio (MGR), a new information theory based hierarchical divisive clustering algorithm for categorical data. MGR implements clustering from the attributes viewpoint which includes selecting a clustering attribute using mean gain ratio and selecting an equivalence class on the clustering attribute using entropy of clusters. It can be run with or without specifying the number of clusters while few existing clustering algorithms for categorical data can be run without specifying the number of clusters. Experimental results on nine University of California at Irvine (UCI) benchmark and ten synthetic data sets show that MGR performs better as compared to baseline algorithms in terms of its performance and efficiency of clustering.