An immune genetic K-means algorithm for Mongolian elements clustering

Abstract

Text clustering is an important area in artificial intelligence. Production of the some character recognitions have been transformed into commercial soft-ware, but the research on Mongolian elements is now just beginning. There are many characters in Mongolian structure and written pattern in contrast with other kinds of characters. In this paper, we proposed a novel clustering technique that combined genetic K-Means algorithm and immune algorithm. The proposed technique clustered the Mongolian elements to the better result. Experiment show that the accurate clustering rate of this method is over 98% and this technique is efficient and feasible.