基于联邦学习的精准广告营销合规分析

Compliance Analysis of Precision Advertising Marketing Based on Federated Learning

摘要: 互联网精准广告营销成为广告市场的主要营销模式，在数据与技术的加持下，精准广告营销给用户和广告主带来了更好的广告体验。精准广告营销作为互联网广告时代发展的必然产物，通过预设用户画像和标签，依靠算法模型进行匹配，在提升广告投放精准度的同时，对用户的数据隐私也带来了不可忽视的风险。国内外针对该问题发布了一系列法律法规，使得用户拥有对其个人数据的掌控权，面对愈发严格的监管，兼顾安全与效率成为广告产业的必然选择。针对这一问题，联邦学习提出新的解决方案，联邦学习可在兼顾用户体验与广告精准度的同时，保护用户的数据隐私与个人信息安全。本文通过论证联邦学习对相关法律规则的供给，对联邦学习精准广告营销场景进行合规性分析，同时提出针对该场景下的监管重点，联邦学习不是完美的解决方案，仍然存在合规挑战。

Abstract: Internet precision advertising marketing has become the main marketing model of the advertising market. With the support of data and technology, precision advertising marketing has brought better advertising experience to users and advertisers. As an inevitable product of the development of the Internet advertising era, precision advertising marketing, by presetting user profiles and labels, relies on algorithm models to match, while improving the accuracy of advertising, it also brings risks to users’ data privacy that cannot be ignored. A series of laws and regulations have been issued both domestically and internationally to address this issue, enabling users to have control over their personal data. Faced with increasingly strict regulations, balancing safety and efficiency has become an inevitable choice for the advertising industry. In response to this issue, federated learning proposes a new solution that can protect user data privacy and personal information security while balancing user experience and advertising accuracy. This article demonstrates the supply of relevant legal rules by federated learning, conducts compliance analysis on federated learning precision advertising marketing scenarios, and proposes that federated learning is not a perfect solution for regulatory priorities in this scenario, and there are still compliance challenges.