Design And Research Of Fetion Monitoring System

Fetion is a hosted business applications and entertainment features instant messaging products launched by China Mobile in the second half of2006.Because of its Economic and convenience, fetion becomes more and more popular. However, the fetion has also been used by criminals. The spread of garbage fetion has affected consumers’normal life.China Mobile, as the three main telecom operators in china, and also the owner of fetion, is obligated to monitor and manage the fetion.Therefore, this paper proposes to establish a fetion monitoring system in the side of the telecom operator.Then the dangers of spam fetion, spam fetion characteristics and governance programs refuse to fetion were elaborated.Based on the analysis of the system architecture and fetion network protocol, the principle of reduction Fetion messages, the paper designed a monitoring systems. Proposed monitoring system includes the fetion central management platform and the monitoring platform. Meanwhile, the filtering technology used in the system were introduced and text classification was taken as a research focus. Finally, to achieve a text classification and conducted a experiments to further validate the text classification technology can be used in the fetion monitoring system.Specific research work is as follows:1. Introducing the basic concepts of spam messages, classification, characteristics and the current domestic situation of spam messages governance and technology, giving the significance and background.2. Considering the transceiver architecture and mechanism of Fetion message system and to study the principles of the restored Fetion messages for the next design fetion monitoring system feasibility analysis. Then introduce kinds of technologies about filtering the fetion.3. Designing detailed system work. Detailed design includes several sub-modules of the system, describes the business processes.4. Reviewing the text classification technology, including key technical classification algorithms basic processes and basic processes, such as Chinese word segmentation, feature extraction and text classification algorithms. Finally achieve a text classification technology.5.Text classification experiment is designed and analyze the results. Further validate the use of text classification in the system accuracy.