Clustering Techniques and Research Challenages in Machine Learning
Ritesh C. Sonawane, Hitendra D. Patil · 2020
Clustering is a technique used to detect a group of similar data. Clustering is generally used for grouping the items with the similar information. Clustering always appears as an unsupervised learning technique. It appears as a common technique used for statistical data, machine learning and computer science analysis. Clustering is an unsupervised learning method that remains suitable for the investigation of interrelationships between samples to make a preliminary assessment of the sample structure. As it is very difficult for human to understand data in a high-dimensional space, clustering is employed. There are many real world applications for clustering. There are several ways to implement clustering algorithm such as centralized, distributed, connectivity, group, graph and density. Existing clustering algorithms are k-means, hierarchical, mean-shift, expectation-maximization, DBSCAN.