A method for estimating the number of clusters in a dataset that is to say flexible for differing types and sizes of data, adjusting to any assembling methods, and smooth to calculate is conferred in this division. A Deviation Ratio Index based on Medoids (DRIM) is the approach we intend. The object distance to the final κ-medoids is promoted to calculate the DRIM method. The block-based κ-medoids treasure (Block-KM) and the κ -medoids constructed utilizing the variance of distance (VarD-KM) were used to acquire these final medoids. Before running the Block-KM and VarD-KM, we select a distinguishing transformation for few datasets. We use ten real datasets to legalize the DRI. These data involve Vote, Soybean (small), Primary Tumor, Breast Cancer, Ionsphere, Iris, Wine, Zoo, Heart Disease Case 2, and Credit Approval dossier. The experimental results show that the DRIM method predicts the number of clusters for the ten evident datasets more precisely than added methods. Three types of pretended data to judge the proposed order resulted in 76.67% of experiments concluding correctly. Applying the new approach to arrangement 62 universities in Indonesia established data on workforce, education, research, institution, infrastructure, and collaboration produces three easily elucidated groups.
Author(s) Details:
Kariyam,
Department
of Mathematics, Faculty of Mathematics and Natural Sciences, Gadjah Mada
University, Indonesia and Department of Statistics, Faculty of Mathematics and
Natural Sciences, Universitas Islam Indonesia, Indonesia.
Abdurakhman,
Department
of Mathematics, Faculty of Mathematics and Natural Sciences, Gadjah Mada
University, Indonesia.
Adhitya Ronnie Effendie,
Department of Mathematics, Faculty of Mathematics and Natural
Sciences, Gadjah Mada University, Indonesia.
Please see the link here: https://stm.bookpi.org/ACST-V5/article/view/12246
No comments:
Post a Comment