The main results of our paper are as follows: present a Chinese text clustering model; obtain a Chinese text algorithm which can select better initial point and a DBTC clustering algorithm which can identify cluster with any shape.

• 本文获得的主要结果：提出了一个中文文本聚类模型，提出一种选取初始聚类中心的中文文本聚类算法和一种能发现任意形状簇的DBTC聚类方法。