The K-nearest-neighbor algorithm is adopted in the classification of the Kazakh text, while in characters chosen, a method that integrates language information and statistical information from the training corpus is applied. The weight of these characters is computed from three parameters: word frequency, centralized degree, decentralized degree. After training, the vector space model of the Kazakh text categorization is got, and the Kazakh text through K-nearest-neighbor algorithm is classified. Experimental results show that this method is feasible.