| 摘要: |
| 研究不同影响力的节点在复杂网络信息传播过程中产生的作用，对于分析舆论传播、预防谣言扩散、引导信息传递等方面具有重要意义。针对传统易感-感染-移出（Susceptible-Infected-Recovered，SIR）模型认为网络中所有节点性质相同，在信息传播中具有相同的接触感染率和恢复率的问题，提出了节点影响力下的改进SIR传播模型——NI-SIR（Node Influence SIR），并对其阈值推导过程展开深度分析。首先，将复杂网络中的节点按照影响力的不同进行分类，不同类别的节点赋予其不同的接触感染率及恢复率，达到模拟真实信息传播过程的目的；其次，对NI-SIR模型的阈值进行推导，从而为进一步判断疾病是否流行或信息扩散的趋势打下理论基础；最后，通过在真实数据集的实验对比，证明NI-SIR模型的偏差率明显低于传统SIR模型，在真实网络中有更好的适用性。 |
| 关键词: 复杂网络 信息传播 SIR模型 传染病模型 节点影响力 |
| DOI： |
|
| 基金项目:国家自然科学基金重点项目（51334003） |
|
| An improved SIR propagation model based on node influence |
| CHEN Ziyang,ZHANG Yuexia |
| (School of Information and Communication Engineering,Beijing Information Science and Technology University,Beijing 100101,China) |
| Abstract: |
| It is important to study the role of nodes with different influences in the process of complex network information dissemination,especially in analyzing the spread of public opinion,preventing the spread of rumors,and guiding the transmission of information.The traditional susceptible-infected-recovered(SIR) model considers that all nodes in the network are of the same nature and have the same contact infection rate and recovery rate in information dissemination.For this problem,this paper proposes an improved SIR propagation model based on node influence(NI-SIR) and derives and analyzes its thresholds.Firstly,the nodes in the complex network are classified according to different influences.Different types of nodes are given different contact infection rates and recovery rates to achieve the purpose of simulating the real information dissemination process.Secondly,the threshold of the NI-SIR model is derived to lay a theoretical foundation for further judging whether the disease is prevalent or the trend of information dissemination.Finally,through the experimental comparison of real data sets,it is proved that the deviation rate of NI-SIR model is significantly lower than that of traditional SIR model,and it has better applicability in real networks. |
| Key words: complex network information dissemination SIR model infectious disease model node influence |
《电讯技术》编辑部 