Application of Self-Organizing Maps in Text Clustering: A Review
Yuanchao Liu, Ming Liu, Xiaolong Wang · InTech eBooks · 2012
Similar as text classification, text clustering is also the technology of processing a large num‐ ber of texts and gives their partition.What is different is that text clustering analysis of the text collection gives an optimal division of the category without the need for labeling the category of some documents by hand in advance, so it is an unsupervised machine learning method. By comparison, text clustering technology has strong flexibility and automatic processing capabilities, and has become an important means of effective organization and navigation of text information. Jardine and van Rijsbergen made the famous clustering hy‐ pothesis: closely associated documents belong to same category and the same request [1]. Text clustering can also act as the basic research for many other applications. It is a prepro‐ cessing step for some natural language processing applications, e.g., automatic summariza‐ tion, user preference mining, or be used to improve text classification results. YC Fang, S. Parthasarathy, [2] and Charu [3] use clustering techniques to cluster users’ frequent query and then the results to update the FAQ of search engine sites.