A Study On Crowdsourcing Quality Assessment Model Based On Blockchain

In recent years,crowdsourcing has become a new business model.As crowdsourcing targets a wide range of groups,the quality of crowdsourcing results is uneven.Therefore,quality control has become an important branch of crowdsourcing research.At present,most of the quality control methods are based on the centralized platform,which cannot guarantee complete credibility.In addition,the incentive mechanism of current crowdsourcing platforms mostly adopts fixed pricing,which leads to the unreasonable distribution of remuneration.In order to solve the above problems,this paper proposes a crowdsourcing quality assessment model based on blockchain,which transfers the power of quality assessment to neither the requester nor the worker,so as to avoid the evil situation of both sides.In this paper,the blockchain idea is introduced to realize a decentralized platform to ensure the security of data and solve the problem of data loss of centralized platform.The credibility mechanism is introduced to construct the credibility model,and the credibility value that accurately and objectively reflects the user's behavior characteristics is taken as a basis for the selection of workers in the crowdsourcing quality assessment.We improve the EM algorithm,solve the sensitivity of initial value of EM algorithm,and implement a crowdsourcing quality assessment method with high accuracy,so as to produce more accurate evaluation results.Finally,based on the results of quality assessment,a two-stage incentive mechanism based on the quality of completion is constructed,so as to provide a more reasonable and transparent way of compensation distribution.Experiments prove the effectiveness of the improved EM algorithm.The credit mechanism can encourage workers to work hard and improve the quality of crowdsourcing results.At the same time,the cost of executing smart contracts on the blockchain can be controlled within a reasonable range.