网络影响力预知模型:一种大数据下高校舆情监测与预警机制

Predicting Model in Network Impact: a Monitoring and Warning System for Public Opinion in Universities under Big Data Framework

互联网对高校大学生的思想传播模式尤其对舆情的传播产生了巨大影响。对于这样的新形势,建立和健全高校大学生舆情监测与预警机制对于及时了解大学生的思想动态,分析和解决思想问题,为大学生树立正确"三观"具有重要的意义。基于此,有必要建立一种监测大学生网络舆情的模型——基于连续时间马尔科夫过程的用户影响力预知模型,通过该模型找到高校社交媒体中最具影响力的用户,锁定最具影响力的用户群(关键人物),采用基于机器学习的自动文本分类方法,对该群体进行分类,主要分为三类:积极型关键人物、中立型关键人物、消极型关键人物,最后针对不同类型的关键人物采取不同的措施以达到对高校大学生社交网络舆情发展的监测与预警。

The Internet has great impact on the dissemination of ideas, and in particular public opinion,among college students. Under these new circumstances, it is of great significance to build up and gradually improve a monitoring and warning system for public opinion in universities, which will enable us to know how the students think, and address relevant issues in order to help them to establish the correct ＂three-values＂. This paper proposes a monitoring system for college student online public opinion, a predicting model of user influence based on the continuous time Markov process, through which we will find the most influential users（key figures）the social network of college students. With an automatic text classification method based on machine learning,the key figures are mainly classified into three categories： positive key figures, neutral key figures, and negative key figures. Finally, the paper proposes some measures in accordance with different types of key figures to promote the development of social networking service for college students.