|
基于OLDA模型的社会热点事件演化路径研究
|
Abstract:
社会热点事件与网络舆情有着紧密联系,已经成为影响社会稳定的主要因素之一,热点事件应变能力反映了网络社会波动的相对稳定能力。本文结合社会热点事件和网络舆情的特征和演化规律,构建社会热点演化指标体系,并对数据进行赋权和标准化等处理,通过灰色关联分析方法得出数据之间的关联度,并进一步根据聚类中心,将社会热点演化等级分为轻微级、一般级和重大级三类,最后利用在线主题模型分析总结事件发展各阶段的主题词,根据主题词推算社会热点事件的演化路径。
Social hotspot events are closely related to network public opinion, which has become one of the main factors affecting social stability, and the adaptability of hotspot events reflects the relative stability of network social fluctuations. Combined with the characteristics and evolution laws of social hotspot events and network public opinion, this paper constructs an index system for the evolution of social hotspots, and processes the data with weighting and standardization, obtains the correlation degree between the data through the grey correlation analysis method, and further divides the evolution level of social hotspots into three categories according to the clustering center: slight, general and major, and finally uses the online topic model to analyze and summarize the topic words at each stage of event development, and deduces the evolution path of social hot spots according to the subject words.
[1] | 余本功, 张卫春, 王龙飞. 基于改进的OLDA模型话题检测及演化分析[J]. 情报杂志, 2017, 36(2): 102-107. |
[2] | 裴可锋, 陈永洲, 马静. 基于OLDA的可变在线主题演化模型[J]. 情报科学, 2017, 35(5): 63-68. |
[3] | 李慧, 王丽婷. 基于话题标签的微博热点话题演化研究[J]. 情报科学, 2019, 37(1): 30-36. |
[4] | 戴长松, 王永滨, 王琦. 基于在线主题模型的新闻热点演化模型分析[J]. 软件导刊, 2020, 19(1): 84-88. |