The Study of an Improved Text Clustering Algorithm for Self-Organizing Maps

Abstract

The traditional SOM algorithm need to determine the number of clustering categories in advance, which is very subjective. In this paper, an improved k-means initial value selection algorithm is proposed to calculate the number of clustering categories, which is applied to SOM network model. In this algorithm, the Latent Semantic Indexing is applied in the pre-processing stage of clustering, and the improved SOM algorithm is applied in the text clustering stage. Namely, the number of clustering categories obtained by the improved k-means initial value selection algorithm is taken as the number of neurons in the output layer of SOM network.

Direct Batch Growth Hierarchical Self-Organizing Mapping Based on Statistics for Efficient Network Intrusion Detection

“…Data clustering can be achieved through the topology representation of the high-dimensional data, that is, the data vectors with similar distribution characters are grouped together and displayed graphically thanks to the visualization advantage of SOM [4]. Thus, SOM naturally has many applications such as image and text segmentation [5], [13], medical diagnosis [6], [12], [14], target recognition [7], and traffic attack detection [8], [9], [20]. However, previous studies have theoretically predicted that the original simple model of SOM fails to fully represent the topology of the asymmetric-distributed (network traffic) data [4], [10].…”