OPTIMASI PSO UNTUK METODE CLUSTERING FUZZY C-MEANS DALAM PENGELOMPOKAN KELAS Max Teja Ajie Cipta Widiyanto

Main Article Content

Redaksi Tim Jurnal

Abstract

Clustering is a method that divides data objects into groups based on information found in data describing objects and relationships between them. In partition-based cluster analysts K-Means method and Fuzzy CMeans Method which is a frequent and commonly used clustering method. However, in its development now with various macar data with the complexity of the variable would be more asked again to the effectiveness and as efficiently whether the method can cluster the data. So it is necessary to optimize where the method has a weakness that is likely to mask the impact of the deficiency in clustering a data object that has a complex variable. Data data used is the value of slot shift one with a variable of the value Academic, nonacademic value in the form of a questionnaire value attitudes assessed by each teacher will be inputted, as a determinant of the division of leading classes in the future in the next teaching. In this study tried to correct some of the deficiencies of the Fuzzy C-Means algorithm ie the selection of early cluster centers and local solutions. An efficient algorithm is proposed to improve the grouping of such methods with Particle Swarm Optimization. In recent years, Particle Swarm Optimization (PSO) has been successfully applied to a number of real-world grouping issues with fast and effective convergence for high-dimensional data.

Downloads

Download data is not yet available.

Article Details

Section
Articles