作　　者： ; (蒙在桥）;

机构地区： 广东工业大学计算机学院

出　　处： 《计算机应用》 2014年第7期1960-1963,共4页

摘　　要： 传统传播模型较难描述在线社交网络中的复杂活跃模式以及节点间的拓扑差异,并且其接触式的传播者退化方式也与现实不符。针对理论模型模拟与现实消息传播的不符,提出一个基于在线社交网络的动态消息传播模型D-SIR。该模型考虑了在线社交网络中影响消息传播的一些实际因素,引入基于传播延迟的退化方式使传播者自发地退化成免疫者,动态指定节点的权威度和免疫力以适应非均质网络,并考虑接收增强信号效应以及外部社会加强效果。在采集的新浪微博真实传播网络数据中,通过参数变化的传播仿真实验验证了D-SIR模型可以有效反映在线社交网络的现实传播情形,并且较传统模型更具灵活性及可扩展性。 Traditional spreading models have difficulties in descripting the complex activity patterns and the topological differences between nodes in online social networks, and the contact-based spreader annihilation mechanisms in these models do not fit with the reality. To filling the gap between spreading simulations of theoretical model and realities of information spreading, a new dynamic information spreading model （D-SIR） based on online social network was proposed. With consideration of some practical factors in information dissemination process, this model introduced the time delay annihilation mechanism that spreaders changed to stiflers spontaneously and the dynamic authority and resistance of nodes mechanism to apply to inhomogeneous networks, and considered the receiving reinforced signal effect and the social reinforcement. With the variances of parameters, the simulations on the real-world online social network which is constructed by crawled Sina microblog data verify that D-SIR model can reflect the real spreading situation in online social network. And compared to the traditional spreading model, the new model is more flexible and extensible.

关 键 词： 在线社交网络 消息传播模型 建模仿真 传播模型 复杂网络

领　　域： [自动化与计算机技术] [自动化与计算机技术]