Comparative Study of K-Means Clustering Using Iris Data Set for Various Distances

K-means clustering is an algorithm, which has been used to cluster the given data into k sets that are mutual exclusive of each other. The K-means algorithm is designed to work with the Euclidean distance but there are many measures to identify the dissimilarity of the dataset. The aim of this paper is to discuss the performance of K-means clustering algorithm on city block, cosine, and correlation distance which are used to get the results and further their performance has been shown in terms of accuracy. For classification, authors have chosen the IRIS data set. K means have claimed 98% accuracy on city block and correlation distance.