Uyghur Words Clustering Based on the Similarity Calculation
Xun Tan · Shinjang dashösi ilmiy jurnili · 2012
Word clustering is oriented to the words of the clustering technique,widely used in natural language processing in all directions.The traditional K-means of the algorithm is based on the distance of the clustering algorithm,the algorithm considers two words the closer,the greater the similarity.This paper puts forward the words based on morphological length and based on similarity calculation of K-means clustering algorithm.Experiments show that,based on the similarity calculation method of morphological results its utility measure E reached 0.555.