Research On The Technology Of Cyberspace User Credit Evaluation

With the development of the Internet, Internet has brought us greatconvenience, while it also causes many problems. These problems notonly make the users feel distress but also increase the cost of the society.It is very urgent to establish an effective cyberspace user credit evaluationsystem to guide the users’ behavior.This paper intend to research on the technology of the user creditassessment for the mainstream network platform. We research on theevaluation index and algorithm to form a complete cyberspace user creditevaluation system which has great significances to build a network creditmonitoring system for our country.First, use process analysis to recognize the key elements when usersusing the network platform, and summarize the credit judge relevantelements. Then establish the general framework for network individualcredit evaluation index system, and research into the detail of cyberspaceusers’credit from four dimensions, including user identification, behavioranalysis, credit association and credit history.Secondly, we choose social network as an example to talk about theexisting credit problems. And use the Dephi method to choose the indexesand give the detail three level index system. In additional, we introducehow the index is related to the user credit. We use sina microblogging as arepresentative of the social network, to illustrate how the raw data is obtained, and the way to calculate the grade of each item based on thequantitative methods. Then we sum up some common used formula forthe grade calculation for later use.After that, we research on the static algorithm to for the creditevaluation system. We use the Analytic Hierarchy Process and FuzzyComprehensive to establish the mathematical model for comprehensiveevaluation. We determine the evaluation level and other relatedparameters, and choose trapezoidal function as the membership function.We take a sina microblogging user as an example to verify theeffectiveness and feasibility of the evaluation algorithm.Last, we study on how social and trust relationship influences theusers’ credit. We also study on the classification and judgment of thedishonest activity and propose a network dishonest activity based usercredit dynamic adjustment algorithm.