基于信号理论和众包的社交媒体平台安全性和可信度群体评估方法研究

The emerging social media now appear significant advantages of comprehensiveness, diversity and wisdom, whereas its security and trustworthiness issue become increasingly serious, as is urgently needed to address. The available studies mainly aim at both social media content and user security, including model, mechanism, protocol and algorithm, lacking of researching on effective, measurable evaluation for security and trustworthiness of social media fundamental platforms and applications, thus heavily influence the further improvement and evolution of social platforms. The study chooses a novel perspective on fundamental platforms in social media ecosystem, creatively introduced Signaling Theory of information management science by cross-disciplinary research methods, so as to explore a generic signaling-based method and theory for crowd evaluation on social media platforms, inspired by the idea of crowdsourcing. Further, the experimental validation and analyses will be accomplished by using a real-world social platform, and also security enhancement mechanism and trustworthiness optimization approach will be proposed to meet the requirement of platform evolution. The study is significantly theoretical meaningful for continuously improving platform security and trust, as well as establishing security-preserving and trustworthy social media ecosystem; it has also better applicable vision and practical application value for realizing secure interaction, sharing and digital rights management of social media content, with promoting healthy, normal and rapid development of digital media content industry.

新兴社交媒体网络正呈现出综合化、多样化和智慧化等显著优势，然而它的安全和信任问题却日趋严重，亟待解决。现有研究工作主要针对媒体内容和用户安全，包括模型、机制、协议和算法等，但仍缺乏对社交媒体基础平台和应用进行有效的、可量化的安全性和可信度评估研究，从而严重地影响到社交平台的进一步改善和进化。本项目研究选择从社交媒体生态系统中的基础平台入手，采用跨学科的研究方法，创造性地引入信息管理科学中的信号理论，并受众包思想启发，探索基于信号的社交媒体平台群体评估一般方法和理论。通过现实社交平台实验验证和分析，提出平台安全性增强机制与可信度优化方法，满足平台进化需求。本研究对不断增强平台安全和可信，构建安全持久、可信赖的社交媒体生态系统，具有重要应用基础理论意义；对实现社交媒体内容的安全交互、分享和数字版权保护，促进数字媒体内容产业的健康、良性和快速发展，也具有较好的应用前景和实用价值。

新兴社交媒体正呈现出综合化、多样化和智慧化等显著优势，然而它的安全和可信问题却日趋严重，亟待解决。现有研究工作主要针对媒体内容和用户安全，包括模型、协议、机制、算法等，仍缺乏对社交媒体平台和应用进行有效的、可量化的安全性和可信性评估研究，从而影响到社交平台的进一步改善和进化。本项目主要研究探索面向社交媒体生态系统的安全可信和群体计算，提出了基于社交情境分析的群体任务分配方法及相关算法，进一步提出了面向社交网络平台安全性和可信性增强的ABE密码学方案、信任评估方法，以及基于机器学习的社交机器人发现方法等。本项目对不断增强平台可信和安全，构建安全持久、可信赖的社交媒体生态系统，具有重要应用基础理论和科学意义。