【摘要】 研究了一种基于SOM(self-organizing map)和层次聚集的中文文本聚类方法,按照提出的中文聚类模型,该方法首先对文档集向量化,文档向量矩阵通过SOM训练映射到虚拟的二维空间,形成初步聚类;然后对虚拟坐标集进行二次聚类.与直接聚类方法相比,该方法提高了聚类的效果,减少了计算时间,通过数值实验对比表明该方法对中文文本聚类具有有效性.更多还原

【Abstract】 This paper studies a clustering method of Chinese documents based on self - organizing map (in short SOM) and Agglomeration. A new Chinese documents model is given. First, Chinese text data are transformed into text vectors, which are used as training data of SOM and mapped by training SOM to a virtual coordinates set that is an initial clustering result for text data. Then the virtual coordinates set is further clustered. It outperforms other algorithms in computation time due to decreasing dimension. Numerical experiment shows that the method is efficient for the clustering Chinese documents.更多还原