Research On Group Intelligence Perception Network Participants And Selection Methods Based On Data Quality

In recent years,with the development of big data and Internet plus technologies,he demand for data in various industries has also evolved into large-scale and complication,which lead the way that ordinary people use smart devices with sensors such as mobile phones to collect perceptual data is popularized,so the Crowd Sensing Network is came into being.Because of its advantage of the unique way of data collection,it is favored by all walks of life,Crowd Sensing Network has already become the hot research topic at present.At present,there are many researches on Crowd Sensing Network,but most of the researches focus on how to perceive the participation of people by material incentives.there is a lack of research on how to choose the perceived participants.In this paper,There are some ways in which participants can choose to improve their data quality.here are two ways to choose participants based on data quality.The main achievements include:This paper proposes a participant selection method based on Knapsack problem and reputation feedback.The method is mainly used to solve the problem of how to choose the participants to maximize the quality of perceived data,when the number of participants is uncertain.This method sets the credibility value of participants' tasks,selects the participants to perform the task through the optimization of knapsack problem,and gives the credit value feedback according to the completion of the task.Finally,experiments prove that the selection method of participaints based on backpack problem and reputation feedback has better effect on the quality of data and the rate of return compared with the commonly used method of selecting participants.This paper proposes a participant selection method based on the improved UCB(Upper Confidence Bound)algorithm.The method is used to solve the problem of how to choose suitable participants to improve the quality of perceived data,in the case of determining the number of participants.By considering the interaction between the quality of perceived data and the perceived data quality in the future,the bias factor about the quality of perceived data is added in the exploration stage,which to guide the whole process to development along the best direction.Effectively balance exploration and exploitation,So as to achieve the goal of selecting the best participants and improving the quality of perceived data.