Study of Unsupervised Learning Algorithms | Original Article

ABSTRACT:

Clustering is the process of grouping a set of objects in such a way that elements in the same cluster are more like each other than datapoints or objects in separate collections or clusters. The partition-predicated clustering approaches discussed in this study include K-Means, K-Means++, and the object-predicated Fuzzy C-Means clustering algorithm. The use of sorted and unsorted data in the algorithms is proposed in this study as a strategy for improving clustering results. The criteria used to examine behavioral patterns include elapsed time and total number of iterations. The experimental results show that sending sorted data instead of unsorted data not only reduces time complexity but also improves clustering approach performance.