A COMPREHENSIVE SURVEY OF K-MEANS CLUSTERING ALGORITHMS: TECHNIQUES, ADVANTAGES, LIMITATIONS, AND ACCURACY IMPROVEMENT STRATEGIES

Abstract

Clustering is the most important stage in the analysis of data, which consists of categorizing the collection of objects into groups referred to as clusters. The most commonly used and essential method of clustering is K-means clustering. It's likewise referred to as "nearest neighbor searching." Several efforts have been taken to boost the accuracy of the K-means clustering method. The elements or investigation points are to be grouped into distinct groups using the k-means approach. The K-means clustering algorithms primary centroids are randomly selected, which poses an important concern because the algorithm's effectiveness largely depends on the initial centroids and may contribute to local modifications. Increasing the clustering outputs of the K-means clustering procedure was the primary aim of the current study. Relevant outcomes from multiple studies have shown the standard clustering technique and the prospects for accuracy improvements in clustering. The present publication provides an overview of the research conducted by several researchers employing K-means clustering. We have additionally expressed regarding the K-means clustering algorithm's benefits and limitations. The article delivers an extensive review of the numerous k-means clustering