An Integrated Clustering Framework Using Optimized K-means with Firefly and Canopies

Abstract

Data Clustering platform is used to identify hidden homogeneous clusters of objects to analyze heterogeneous data sets based upon the attribute values in the domain of Information Retrieval, Text Mining, Web Analysis, Computational Biology and Others. In this work, a hybrid clustering algorithm for K-Means called Optimized K-Means with firefly and canopies, has been proposed by integration of two meta-heuristic algorithms: Firefly algorithm and Canopy pre-clustering algorithm. The result model has been applied for classification of breast cancer data. Haberman’s survival dataset from UCI machine learning repository is used as the benchmark dataset for evaluating the performance of the proposed integrated clustering framework. The experimental result shows that the proposed optimized K-Means with firefly and canopies model outperforms traditional K-Means algorithm in terms of classification accuracy and therefore can be used for better breast cancer diagnosis.