基于移动社交大数据的复杂社会网络行为传播扩散与预测关键技术

With the rapid development and applications of the emerging technologies in the mobile Internet environment, the social network presents new features such as the initiative, self-organization, heterogeneity, and the anonymity, randomness and subjectivity of the interaction between individuals. The conventional information diffusion, forecasting methods can not satisfy the requirements of real-time and reliability of complex social networks. To address these problems, this project investigates the big data-driven management and decision process based key technologies of complex social network behavior diffusion and prediction based on mobile social data. First of all, with the establishment of general correlation model, the research of behavior diffusion and forecasting strategies are studied. Secondly, by using the mathematical tools such as game theory, operations research and optimization algorithm, this project studies the variation characteristics of user in propagation process, the diffusion strategy based on collaboration and incentive, and the forecasting strategy based on dynamic topology and network evolution. By fully investigating the users’ behavior characteristics and establishing a reasonable allocation solution of network resources, it can achieve the improved data dissemination efficiency, diffusion control and the network performance prediction. The development of this project can provide theoretical guidance and technical support for the study of the diffusion and prediction of complex social network behavior for big data-driven management and decision-making process.

随着移动互联环境下的新兴技术快速发展与应用，社会网络呈现出参与主体的能动性、自组织性、异质性，以及个体间信息交互的匿名性、随机性和主观性等新特征，使得传统的行为传播扩散与预测方法无法满足当前复杂社会网络对实时性和可靠性的需求。为克服上述问题，本项目基于移动社交大数据，面向大数据驱动的管理与决策，研究复杂社会网络行为传播扩散与预测的关键技术。首先，通过建立通用关联模型为行为传播扩散与预测策略研究奠定基础。其次，利用博弈论、运筹学、优化算法等数学工具，分别研究基于用户状态变化特性的传播策略、基于协作和激励的扩散策略、基于动态拓扑结构和网络演化特性的预测策略。通过充分挖掘用户行为特征、合理分配网络资源，以达到提高传播效率、扩散可控性、网络预测性能的目的。本项目的开展可为面向大数据驱动的管理与决策的复杂社会网络行为传播扩散与预测的研究和应用提供理论指导与技术支撑。

随着移动互联环境下的新兴技术快速发展与应用，社会网络呈现出参与主体的能动性、自组织性、异质性，以及个体间信息交互的匿名性、随机性和主观性等新特征，使得传统的行为传播扩散与预测方法无法满足当前复杂社会网络对实时性和可靠性的需求。为克服上述问题，本项目基于移动社交大数据，面向大数据驱动的管理与决策，研究复杂社会网络行为传播扩散与预测的关键技术。首先，通过建立通用关联模型为行为传播扩散与预测策略研究奠定基础。其次，利用博弈论、运筹学、优化算法等数学工具，分别研究基于用户状态变化特性的传播策略、基于协作和激励的扩散策略、基于动态拓扑结构和网络演化特性的预测策略。通过充分挖掘用户行为特征、合理分配网络资源，以达到提高传播效率、扩散可控性、网络预测性能的目的。本项目的开展可为面向大数据驱动的管理与决策的复杂社会网络行为传播扩散与预测的研究和应用提供理论指导与技术支撑。