Comparison Of Clustering Levels Of The Learning Burnout Of Students Using The Fuzzy C-Means And K-Means Methods

Abstract

Learning burnout is an impact from work done in a manner Keep going continuously, causing fatigue physical and emotional. If boredom study no handled, got cause students no productive and inhibits potency student . So from that study this proposed method clustering for group level saturation study students. The clustering process in research this use Fuzzy C-Means and K-Means. According to the previous study, Fuzzy C-Means and K-Means can produce results in the best clusters. Destination of study this is to compare performance from method Fuzzy C-Means and K-Means. The dataset used in this study is the boredom of students. Testing was conducted with the use amount clusters 3,4,5. Test results system with method Fuzzy C-Means get score Meanwhile, the global silhouette coefficient is 0.278 for K-Means results testing get score The global silhouette coefficient is 0.287. Temporary for results Davies Bouldin Index, methods Fuzzy C-Means get score 0.224and the K-Means method get value 0.384 of value, the Fuzzy C-Means generates more clusters _ good from K-Means. However both of them have weak structure _ because some data has data distance between one more clusters far from distance between different data clusters, so that creates that data worth.