首页|期刊导航|网络安全与数据治理|反诈场景词元资产化建模与多主体协同价值分配机制

反诈场景词元资产化建模与多主体协同价值分配机制OA

Anti-fraud token assetization modeling and multi-agent collaborative value distribution mechanism

中文摘要英文摘要

面向反诈协同治理中的数据壁垒、价值量化和激励失衡问题,构建了多险种兼容的反诈业务价值词元体系,提出融合精准度、时效性、稀缺性、风险等级与实战增益的资产化估值方法,量化警银网通异质数据贡献.建立多保险主体协同收益池,设计按保费受益比例的差异化反哺机制,并通过省级季度仿真验证其有效性与稳健性.结果表明,该机制可准确区分异质数据贡献,实现激励相容分配,为缓解反诈保险高赔付、推广难和主体权责利不匹配提供市场化路径,推动反诈治理由财政驱动向市场造血转型.

To address data barriers,value quantification difficulties,and incentive imbalances in anti-fraud governance,this paper constructs a multi-insurance-compatible business value token system and proposes an asset valuation method integrating accuracy,timeliness,scarcity,risk level,and practical gain to quantify heterogeneous data contributions from police,banks,internet platforms,and telecom operators.A col-laborative benefit pool and a premium-based differentiated feedback mechanism are established to link governance value with market returns.A provincial quarterly simulation calibrated with public statistics verifies the model's effectiveness and robustness.Results show that the mecha-nism accurately distinguishes heterogeneous contributions and achieves incentive-compatible distribution,offering a market-oriented path for mitigating high claim costs,promotion difficulties,and responsibility-benefit mismatches,and supporting the shift from fiscal-driven to market-sustained anti-fraud governance.

王先兵

武汉大学 国家网络安全学院,湖北 武汉 430072

信息技术与安全科学

反诈治理数据要素词元经济资产化建模多主体协同价值分配机制市场化闭环

anti-fraud governancedata elementstoken economyasset modelingmulti-subject collaborationvalue distribution mecha-nismmarket-oriented closed loop

《网络安全与数据治理》 2026 (9)

26-33,8

10.19358/j.issn.2097-1788.2026.09.004

评论