摘要
针对文本聚类所面临的维数灾难、稀疏向量以及标准K-Means算法初始中心点选择的随机性等问题,提出了一种基于概念格的文本聚类算法,且该算法不需要评价函数。实验结果验证了该算法的有效性。
It is still difficult to deal with the dimension catastrophe, the sparse vector, and random selection of initial center in standard K-Means algorithm. A new clustering method based on concept lattice without evaluation function was proposed in this paper. Finally, an experiment was given. The results clearly show the outstanding performance of the proposed method in terms of correctness and efficiency.
出处
《计算机应用》
CSCD
北大核心
2008年第9期2328-2330,2334,共4页
journal of Computer Applications
基金
赣教技字(2007)208号