سال انتشار: ۱۳۸۶

محل انتشار: اولین کنفرانس داده کاوی ایران

تعداد صفحات: ۸

نویسنده(ها):

Mohamd Reza Kangavari – Faculty of Computer Engineering Iran University of Science and Technology, Tehran, Iran
Babak Fakhar – Islamic Azad University of Mahshahr

چکیده:

Clustering is grouping of patterns according to similarity in some perspectives. Various data representations, similarity measurements and
organization manners, have made several classes of clustering methods that each one can be a strong method in its own field. Some recent
researches show that ant colony optimization algorithms have been successfully applied to combinatorial optimization problems. In this
paper, we present a new data clustering method for data mining in large databases based on Ant Colony Optimization Algorithm. We adopt
simulated annealing concept for ants to decreasingly visit the number of cities to get local optimal solutions. Our simulation results show
that the proposed novel clustering method performs better than the Genetic K-Means Algorithm (GKA). In additional, in all cases we studied, our method produces much smaller errors than the GKA.