Network Text Proactive Detection Analysis System

With the rapid development of Internet and wide application, the network has become the people to receive information dissemination and one of the major carriers.Network version of the most important network information as a form of expression in a variety of benefits to mankind, but also a variety of human harmful information, such as spam, reactionary information, pornography, superstition information, fraudulent information. These harmful information has seriously affected the normal production and human life, to society caused tremendous economic loss and social harm.In response to the reality of the situation, this article initially on the existing network of text classification detection technology, summed up the overall detection of text classification process, to discuss the existing text classification detection technology applications in information security deficiencies, and then on this basis, the text of the initiative put forward a network analysis system for testing the overall framework, the use of training to learn to initialize the text using the text of the training module extracts text features, text classification by the detection module to quickly and efficiently detect, identify the results of real-time reporting, dynamic self-learning modules in the user automatic learning of artificial classification of the participation of test results, dynamic update systems in order to achieve self-renewal and improvement.Finally, to test the target of spam, the system tests, the experimental results show that: the system can achieve real-time spam detection classification, a higher accuracy, to the text of information security for the network play the role of advice and guidance to achieve to avoid and reduce the loss of purpose.