Design And Implementation Of Anti Cheating System Based On Analysis Of Content And User Behavior About Social Platform

With the booming of the Internet, unprecedented changes have happened in the way people live. More and more people are beginning to rely on the efficiency and vast amounts of information through the internet. With the rapid development of the Internet and the surge in the number of Internet users, the network has become a more appropriate platform to disseminate the effect of the various online advertising. In order to attract more users to focus on their products, some advertiser even advertise information to the social network platform, thus cause the product quality issues, so the goal of this anti-cheating system is to reduce social network platform cheat information rate, thereby to improve the quality of related products and satisfy the user’s requirement.The data processing flow of this system can be divided into four modules: incrementally obtain the data submitted by the product user during the latest seconds, call the anti-cheating strategies proposed by this project, score the new data and then deal with the submitted data according to the score.The implementation of this system includes five parts, and the first part is the development of data extraction platform based on the incremental calculate method, The second part is the development of the anti-cheating text dimension strategy based on the support vector machine modeling method, The third part is the development of the anti-cheating user dimension strategy based on the logistic regression modeling method, The forth part is the development of the anti-cheating user relationship dimension strategy based on the density-based spatial clustering of application with noise modeling method and the last part is the development of anti-cheating strategy monitoring platform based on the statistical method. Thus these five parts cover the whole procedure of the anti-cheating system which ensure the strong integrity and perfection of this anti-cheating system.The implementation of this system testing uses the use case based testing method to verify the functional requirement, and especially as for the testing of the user dimension and text dimension strategy, besides use the use case table to verify the functional requirement,it also use the statistical method to verify the precision rate and recall rate by randomly extracting and scoring the data that submitted by the product user. Through these methods, it has proved that the system has reached the intended target. Besides those user data that this system strategy can’t covered or misjudged, this paper has carried out a detailed analysis and proposed the future improvement of this program.By running the comprehensive test of the system, it has verified that this system could achieve the detection and remove the cheating data, besides, the strategy monitoring part also has reached the target of alarming the unexceptionable bug to ensure this anti-cheating system platform can run smoothly.