Research On Insurance Data Sharing Incentive Mechanism And Privacy Protection Based On Blockchain

Insurance technology is currently in an important stage of empowering insurance,and its widespread application will have a huge impact on the future development of the insurance industry.Insurance technology advocates that insurance companies conduct Digital transformation through data sharing to promote high-quality development of insurance companies.At present,there are problems in internal data sharing among local insurance companies,such as low participation of branch offices in sharing and high risk of privacy leakage.Blockchain technology provides new ideas for the research of secure data sharing between insurance company branches.The main research work of this article is as follows:(1)We have constructed a blockchain based privacy protection insurance data sharing framework.The framework includes an incentive process for insurance data sharing,on chain insurance data sharing process and off chain insurance data privacy protection processing process.The comparison with traditional data sharing frameworks and blockchain based data sharing frameworks demonstrates the role and advantages of this framework in promoting data sharing between branch offices and protecting data privacy.(2)To solve the problem of low participation in data sharing among branches,an incentive mechanism for insurance data sharing based on evolutionary game theory is designed.Firstly,a mutual evolutionary game data sharing incentive model(Insurance Data Sharing-Incentive Model,IDS-IM)between insurance branches is constructed to analyze the stable state that ultimately evolves under different initial conditions;Secondly,parameter assignment simulation experiments were conducted to verify the correctness of the stable state during the policy evolution process,ensuring that the parameter value range meets the conditions of data sharing strategy.Prove that the proposed Insurance Data Sharing Incentive Model(IDS-IM),compared to the EGDSI incentive model,can more quickly incentivize various branches to participate in sharing,ensuring a data sharing participation rate of over 95%.Finally,an incentive mechanism(Evolutionary Game Theory dynamic Incentive,EGTDI)was designed based on the triggering conditions of the evolutionary game data sharing model incentive,and an incentive smart contract designed based on the evolutionary stability strategy was implemented on the Hyperledger Fabric framework.(3)A data sharing algorithm based on improved DP-K-prototypes(Improve Kprototypes-Differential Privacy,IKP-DP)is proposed to address the issue of reduced data availability when differential privacy processing is applied to insurance datasets with mixed attributes.The K-prototypes clustering algorithm was improved by using the difference measurement formula for numerical and categorical attributes,and the selection of initial clustering centers was optimized by combining the average difference.After clustering,differential privacy protection is applied to the cluster center of each cluster.Integrate the proposed IKP-DP algorithm with the standard difference privacy algorithm,(Maximum Distance to Average Vector,MDAV)algorithm,and DP-K-prototypes algorithm in experimental comparisons on the Health＿Insurance Claims dataset,using SSE to evaluate data availability and RL to evaluate information leakage.The results show that the algorithm significantly improves data availability compared to the standard difference privacy algorithm and MDAV algorithm while ensuring minimal privacy leakage.Compared with the DP-Kprototypes algorithm,the data availability also improves with increasing clustering.(4)We have designed and implemented a blockchain based insurance data sharing prototype system.The Hyperledger Fabric alliance chain is selected as the underlying framework,and the Go language is used on the Hyperledger Fabric framework to implement data uplink,data query and incentive mechanism smart contracts.The implementation process of smart contracts is described,and the key codes for adding branches to participate in the game function Add Participant,the activation function EGTDI,the upload information function add Data,and the query information function query Datas By Id are respectively described in detail.Showcased the functions of user management,branch department management,insurance data provision record upload,insurance data usage record upload,insurance data provision record query,insurance data usage record query,insurance data summary record query,and insurance data summary upload of the prototype system.