An Improved Clustering Algorithm Based on k-Means and Artificial Bee Colony Optimization for Datasets that Contain Outliers
Anu Balachandran, K. A. Abdul Nazeer · 2018 International Conference on Computing, Power and Communication Technologies (GUCON) · 2018
k-Means clustering algorithm is the most widely used algorithm in clustering. It is most popular because of its simplicity. There are a lot of issues faced by k-Means algorithm such as, low quality of clusters formed, inability to detect outliers and solutions that can be local optimal solution. In this paper a simple outlier detection algorithm that makes use of Mean and Standard Deviation, is applied on datasets. These datasets are then given as input to an already existing hybrid clustering algorithm of k-Means and Artificial Bee Colony(ABC) algorithm. By applying the outlier detection algorithm, it ensures that the clusters formed are of better quality.