基于概率主题模型的京津冀协同发展研究主题演化分析

京津冀协同发展是当前各学科领域共同关注的热点话题。为深入了解京津冀协同发展话题的研究趋势,以2009～2019年中国知网收录的14 235篇与京津冀协同发展相关的期刊论文为研究对象,采用概率主题模型方法,提出以困惑度、主题平均相似度为指标确定最优主题数,结合文献发表时间挖掘期刊论文潜在主题,从主题强度和主题相似度等多个角度分析主题演化趋势。通过数据分析,挖掘10个潜在主题,生成主题强度年度变化趋势,构建主题内容演化路径,分析主题演化规律。以期为深刻认识京津冀协同发展和科学决策提供理论依据。

An Analysis of Topic Evolution of Coordinated Development of Beijing-Tianjin-Hebei Based on Probabilistic Topic Models

The coordinated development of Beijing-Tianjin-Hebei is a hot topic in various fields. In order to deeply understand the research trend on the topic of coordinated development of Beijing-Tianjin-Hebei, a probabilistic topic model was adopted to analyze the 14, 235 journal papers in CNKI collected from 2009 to 2019. The optimal number of topics was determined by taking the degree of perplexity and the average similarity of topics. Based on the time of publications, this paper explores the potential topics of journal papers, and analyzes the trend of topics evolution from the perspectives of topic intensity and topic similarity. Through data analysis, 10 potential topics are developed, annual variation trend of topic intensity is generated, evolution path of topic content is constructed, and topic evolution rules are analyzed. This paper tries to provide theoretical basis for understanding the coordinated development of Beijing-Tianjin-Hebei and making scientific decisions.