Evolution analysis of online topics based on ‘word-topic’ coupling network

Abstract

Analyzing topic evolution is an effective way to monitor the overview of topic spreading. Existing methods have focused either on the intensity evolution of topics along a timeline or the topic evolution path of technical literature. In this paper, we aim to study topic evolution from a micro perspective, which not only captures the topic timeline but also reveals the topic status and the directed evolutionary path among topics. Firstly, we construct a word network by co-occurrence relationship between feature words. Secondly, Latent Dirichlet allocation (LDA) model is used to automatically extract topics and capture the mapping relationship between words and topics, and then a ‘word-topic’ coupling network is built. Thirdly, based on the ‘word-topic’ coupling network, we describe the topic intensity evolution over time and measure topic status considering the contribution of feature words to a topic. The concept of topic drifting probability is proposed to identify the evolutionary path. Experimental results conducted on two real-world data sets of “COVID-19” demonstrate the effectiveness of our proposed method.

Multimodality Fusion based Topic Detection and Evolution Analysis of Web Videos

“…In recent years, some scholars believe that topic evolution can be analyzed from multiple dimensions such as structure, intensity and content [39][40]. On this basis, Zhu et al proposed a 'word-topic' coupling network to analyze the process of topic [41] in order to comprehensively describe the evolution process and state of topics.…”

Evaluating Advancements in Forensic Medicine and Toxicology Through a Scientometric Mapping

“…It can help to identify the most important topics and themes in a field, as well as the most prominent authors and institutions. Keyword centrality can also help to identify the relationships between different topics and themes in a field and to identify trends and changes over time (19).…”

Spatial-temporal evolution pattern and optimization path of family education policy: An LDA thematic model approach

“…The LDA topic model is not limited by the size of the text, effectively minimizing human intervention and yielding a document-topic matrix and topic-vocabulary matrix [ 39 ]. The probability of each topic appearing in the document can reflect the degree of importance a policy document assigns to a topic, thereby indicating the relative importance of the topic [ 30 ].…”

“…Compared to traditional manual coding, automated analysis methods based on big data offer significant advantages in dealing with extensive and lengthy texts [ 29 ]. The Latent Dirichlet Allocation (LDA) topic model is particularly dominant in this context [ 30 ]. Machine learning-based text analysis method have been widely used in policy research, such as the studies of digital government construction [ 31 ], and the interactive analysis of network public opinion and government response [ 32 ].…”