摘要: 提出了一种新的动态模糊自组织神经网络模型(DFKCN),并将其用于文本聚类中。将DFKCN模型应用到中文文本聚类中,该文的特征向量的表示采用隐含语义分析理论,以体现特征词的语义关系并实现特征词的降维。仿真表明本聚类法的精度高于TGSOM算法的精度,算法的执行效率高于TGSOM,取得了良好的效果。
关键词:
模糊自组织神经网路,
模糊C均值,
聚类数,
文本聚类
Abstract: This paper proposes a new model of dynamic fuzzy kohonen neural network(DFKCN) which is applied to the text clustering. The model DFKCN is used in Chinese text clustering. The text eigenvector is represented by using the latent semantic analysis (LSA), which embodies the semantic relation of the eigen words, and realizes the dimension reduction of the eigenvector. The experiment shows both the precision and the efficiency of clustering DFKCN are higher than those of TGSOM.
Key words:
Fuzzy kohonen neural network,
Fuzzy C means,
Number of clustering,
Text clustering
中图分类号:
耿新青;王正欧. DFKCN:一种动态模糊自组织神经网络及其应用[J]. 计算机工程, 2006, 32(20): 22-24,4.
GENG Xinqing; WANG Zhengou. DFKCN:A Dynamic Fuzzy Kohonen Neural Network and Its Application[J]. Computer Engineering, 2006, 32(20): 22-24,4.