An improved K-Means clustering algorithm: A step forward for removal of dependency on K

K-Means is one of the most popular partition based clustering technique. K-means has gain popularity because of its simplicity and speed of classifying massive data rapidly and efficiently. However, the output of K-Means algorithm highly depends upon the selection of initial cluster centers because the initial cluster centers are chosen randomly. The other limitation of the algorithm is to input the required number of clusters. This requires some sort of intuitive knowledge about appropriate value of K which is sometimes difficult to predict as it requires domain knowledge. In this paper, we have proposed an algorithm based on the K-Means, but it does not require the number of clusters K as input. The time complexity and quality of the clusters produced by the proposed algorithm is compared with that of original K-Means using two different data sets.