基于机器学习的中文文本自动分类的研究和应用-计算机软件与理论专业论文.docxVIP

武汉理I：大学硕士学位论文Abstract 武汉理I：大学硕士学位论文 Abstract With the fact that the science and technology swiftness and violence develops，especially,the Intemet fleetness develops，information is voluminous，the upper Chinese of Interact Web page information number counts witll hundred million on the net．The relevance information how to make use ofrapidly effective gaining of computer art already is that one of computer processing of chinese language field studies a hot spot．Most network search engine is to seek amount of work needed information by self according to having built all index to subjective information analysis，having lightened people thereby to a certain extent mainly．Come to say search engine is partly artificial way of semi·automatization as before to network information arranging right away at present．Intellectualized degree，rise the information how to improve search engine searches for search for efficiency，the amount ofwork reducing the manpower repeated reeonnaissanco already becomes the current intelligence reconnaissance’S to study a hot spot．The Chinese version classification technology is similar to problem for this purpose have provided practical feasible one solving approach． Main automation classification method the main body of a book is introduced first，analyses Chinese WEB text characteristic and brings forward Chinese WEB text automation classification key technology，population design and every function module having realized one kind of Chinese WEB text automation classification implement．Automatic classification of Chinese texts based On machine learning comprises three major function modules，i．e．，web spider,web text preprocessing and web text automatic classifier．The traversing strategy of web spider，the integrated HTML Parser of SUN Mierosystems lnc．，and WEB text filtering technology are firstly discussed．Based On the discussion，the method of word segmentation in Chinese texts and the storage structure of Chinese dictionary,and the method