Collaborative Spam Filtering Model Based On Social Network

At present,the number of spam has exploding,and forms changed.To improve the efficiency and quality of spam filtering,and change traditional spam filtering methods which rely on information feature extraction,and explore the collaborative filtering method using social network.Not only can the social problems brought by garbage information be reduced,and improve the user experience,but also is one of research hotspots of information security technology.The article aims to propose an intelligent model,On the one hand,which is able to get rid of disadvantages of traditional filtering model relies on information attributes,and can enhance dynamic adaptability of the mode on the other.Collaborative spam filtering model based on social network mainly uses social networking community features,and bond with feedback-update-method to establish efficient system for spam filtering mechanism.Based on analyzing the content and forms of garbage of SMS and common timely communication software,as well as the social network as a community divided into theoretical basis of small-world and scale-free properties,puts forward the k-means clustering algorithm based on multi-objective collaborative recommendation framework for social networking community division,at the same time,combining with community members grouping importance to improve the quality of spam filtering.In this paper,the test data and complete evaluation system are designed,and use the data to write the corresponding program of the model was tested.Results show that the model is proposed by this article realized spam filtering,and compared with other spam filtering algorithm also shows that the model in spam filtering has been improved the success rate significantly.