Fuzzy clustering has become a widely used data mining technique and plays animportant role in grouping, traversing and selectively using data for userspecified applications. The deterministic Fuzzy C-Means (FCM) algorithm mayresult in suboptimal solutions when applied to multidimensional data inreal-world, time-constrained problems. In this paper the Quantum-behavedParticle Swarm Optimization (QPSO) with a fully connected topology is coupledwith the Fuzzy C-Means Clustering algorithm and is tested on a suite ofdatasets from the UCI Machine Learning Repository. The global search ability ofthe QPSO algorithm helps in avoiding stagnation in local optima while the softclustering approach of FCM helps to partition data based on membershipprobabilities. Clustering performance indices such as F-Measure, Accuracy,Quantization Error, Intercluster and Intracluster distances are reported forcompetitive techniques such as PSO K-Means, QPSO K-Means and QPSO FCM over alldatasets considered. Experimental results indicate that QPSO FCM providescomparable and in most cases superior results when compared to the others.
展开▼
