An Enhanced Bat Algorithm for Data Clustering Problems

Abstract

Data Clustering in Data Mining is a domain which never gets out of focus. Clustering a date was always an easy task, but achieving the required accuracy, precision and performance have been never so easy. K means being an archaic clustering algorithm got tested and experimented thousands of times with a variety of datasets due to its robustness and simplicity but what this algorithm proposed was not suggested before. The proposed algorithm uses K means Algorithm for the Evaluation and Validation purposes whereas Optimization of the data is done by the help of Bat Algorithm. The drawbacks of K mean mainly its local convergence and initializing number of clusters at the early stage, which are still an issue has aroused the process of working on this algorithm. So for attaining the global convergence the Swarm Intelligence is preferred over Genetic Algorithm and many other techniques. For the latter one the algorithm combined two functions one of them help in knowing the number of clusters which are optimal for the particular dataset and the other one validates the results using another function and compares the various metrics which will define the goodness and fitness of the algorithm. In one line the complete overview of the proposed algorithm can be described, performing validation with the help of a numerical function of the k means and giving the final touch of Optimizing the data by k means bat algorithm’. The algorithm is tested for over 4 datasets available in UCI Repository and the results were expectedly great. Keywords: Data mining · Data clustering · Optimization · Bat Algorithm Data Mining, K means Algorithm