ACM Computing Surveys (CSUR)
A Modified Version of the K-Means Algorithm with a Distance Based on Cluster Symmetry
IEEE Transactions on Pattern Analysis and Machine Intelligence
Multiobjective Evolutionary Algorithms: Analyzing the State-of-the-Art
Evolutionary Computation
Classification and Learning Using Genetic Algorithms: Applications in Bioinformatics and Web Intelligence (Natural Computing Series)
Statistical Comparisons of Classifiers over Multiple Data Sets
The Journal of Machine Learning Research
GAPS: A clustering method using a new point symmetry-based distance measure
Pattern Recognition
Computational Geometry: Algorithms and Applications
Computational Geometry: Algorithms and Applications
A fast and elitist multiobjective genetic algorithm: NSGA-II
IEEE Transactions on Evolutionary Computation
An Evolutionary Approach to Multiobjective Clustering
IEEE Transactions on Evolutionary Computation
A Simulated Annealing-Based Multiobjective Optimization Algorithm: AMOSA
IEEE Transactions on Evolutionary Computation
Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images
IEEE Transactions on Pattern Analysis and Machine Intelligence
Engineering Applications of Artificial Intelligence
Hi-index | 0.10 |
In this paper, we have proposed a multiobjective clustering technique which optimizes simultaneously two objectives, one reflecting the total 'goodness' present in the data set in terms of total compactness (measured using Euclidean distance) of the clusters, and the other reflecting the total symmetry present in the clusters of the data set. The proposed algorithm uses a simulated annealing based multiobjective optimization method as the underlying optimization criterion. Center based encoding is used. The proposed multiobjective clustering technique is able to suitably evolve these cluster centers in such a way so that the two objectives are optimized 'simultaneously'. Assignment of points to different clusters is done based on the newly developed point symmetry based distance rather than the Euclidean distance. Results on eight artificial and six real-life data sets show that the proposed technique is well-suited to detect true partitioning from data sets with clusters having either the hyperspherical shape or point symmetric structure. Results are compared with those obtained by five existing clustering techniques, one multiobjective clustering technique, MOCK, average linkage clustering algorithm, expectation maximization clustering algorithm, well-known genetic algorithm based K-means clustering technique (GAK-means) and a newly developed genetic algorithm with point symmetry based clustering technique (GAPS).