Graph-Based Trajectory Visualization for Text Mining of COVID-19 Biomedical Literature

• Activity Number: 384 - Next-Generation Sequencing and High-Dimensional Data
• Type: Contributed
• Date/Time: Thursday, August 12, 2021 : 12:00 PM to 1:50 PM
• Sponsor: Biometrics Section
• Author(s): Yeseul Jeon* and Dongjun Chung and Jina Park and Ickhoon Jin
• Companies: Yonsei University of Applied Statistics and Ohio State University and Yonsei University of Applied Statistics and Yonsei university
• Keywords: Topic Visualization; Latent Space Item Response Model; Biomedical Literature Mining; Trajectory Visualization; Topic Modeling

Abstract

Upon the emergence of the worldwide pandemic of COVID-19, relevant research has been published at a dazzling pace. It is practically impossible to implement this task manually due to the high volume of the relevant literature. The topic modeling has been considered to be a powerful approach to address this challenge. However, in spite of its potential utility, the results generated from this approach are often investigated manually. In order to address these challenges, we propose a novel analytical framework for effective visualization and mining of topic modeling results. Specifically, with topic-words distributions obtained from the Biterm Topic Model, we model these latent topics as networks to visualizes the relationships among topics. Moreover, the proposed approach allows tracing the change of relationships among topics by plotting the trajectory plot when the input of words is different. Therefore, we could define the fundamental positions of topics which helps to figure out the overall research. The application of this analytical framework to the PubMed literature shows that our approach facilitates understanding of the topics constituting the COVID-19 knowledge.