Clustering or the unsupervised classification of data items into clusters can reveal some intrinsic structures among data. The intrinsic structures like the number of clusters are key issues of data mining. In this paper we propose some genetic guided model based clustering techniques to determine the optimal number of clusters and the characteristics of these clusters automatically. The model based clustering techniques are used to describe the clusters and tune their descriptions while genetic algorithms lead the search to some promising search space Several clustering-specific genetic operators are developed to enhance the search procedure The simulation results on both synthetic and real life data sets demonstrate that our proposed clustering techniques outperform two widely used model based clustering algorithms.
|Title of host publication||Proceedings of the International Conference on Artificial Intelligence : IC-A!'2001|
|Editors||H. R. ARABNIA|
|Publication status||Published - Jun 2001|
|Event||International Conference on Artificial Intelligence - Las Vegas, United States|
Duration: 25 Jun 2001 → 28 Jun 2001
|Conference||International Conference on Artificial Intelligence|
|Period||25/06/01 → 28/06/01|