基于多标准信誉模型的垃圾语音检测-通信与信息系统专业论文.docxVIP

华 华 中 科 技 大 学 硕 士 学 位 论 文 II II Abstract With the development of communication technology, mobile phone users are increasing rapidly. Although people enjoy the convenience of this technology, cell phone gradually become the target of spam calls because of its personal and privacy feature. According to survey, more than 90% people are exposed to the harassment of spam calls, the surfeit of spam calls are annoying people’s daily life while wasting much bandwidth resource. On the condition that we lack enough bylaws to impose sanctions to those spam-call-senders, it’s a better choice to use technological methods to weaken those spam calls. There are many methods have been proposed to detect spam-calls, such as black-and-white list, Turing test, prepaid fees, these methods have already shown some effects. But more and more spam-calls are human issue, they can bypass the machine detection; they are too cunning to detect. So people begin to focus on the reputation system. The research on reputation system have already achieved good results, but there are still some issues haven’t been taken into account: 1) How to deal with multi-standard criteria? 2) How to accelerate learning speed? 3) How to make filter model more flexible when users’ behavior is changing? To solve those problems above, this paper proposes a multi-standard reputation model based on group learning solution. The user to be classified according to their disposition to trust solved the multi-model evaluation criteria problem; group learning mechanism has been established after dividing users into different small groups, so the model processing speed will not slow down because of the large amounts of data; besides, with the stream data processing methods and timing of the sliding window model update mechanism ensures that the model is able to change according to the users changes to ensure the accuracy of the model. Simulation results show that the introduction of dividing trust tendency makes t