Research On Text Data Analysis And Filter Technology Of Campus Network Based On Vector Machine

With the rapid development of Internet, information resources in the network are increasing with astonishing speed. Internet has been the important information vehicle and one of the major channels of transmitting information around the word. Meanwhile, Internet provides massive information resource for people, it also has very tremendous negative impacts on network users. The destruction of malicious virus, the spread of illegal information, the leakage of security information and garbage email in the network are becoming prominently threatens to network. Therefore, the researches into data analysis of the network and filter to illegal information or not meeting requirements in the network make important theoretical and practical significance for improving benign development and further construction of Internet.Based on the introduction to the essential technology of data analysis and filter technology, this thesis presents a general model and arrangement plan of the data analysis and filter system on campus network based on the vector machine. It also gives a profound discussion on support vector machine, text categorization technology, text data analysis and filter technology based on the vector machine.In the text data analysis and filter, the traditional technology is analysis and filter which mainly aims at the address and the port as well as match method based on the keywords. Although these technologies have satisfied the request on the information security filter to great degree, its accurate rate is not high. The study proposes the method with which text data analysis and filters based on support vector machine can overcome the above limitations.To sum up, this study addresses the dictionary model structure, the text data classification, the vector space production method of support vector machine and the text fast vectorization method involved in text data analysis and filter technology based on support vector machine. In addition, it also solves the issue of dictionary model structure and text fast vectorization method regarding the text data analysis and filter on the basis of the vector machine.