Citation Graph Analysis and Alignment Between Citation Adjacency and Themes or Topics of Publications in the Area of Disease Control Through Social Network Surveillance

• Published in: Disease Control Through Social Network Surveillance pp. 89 - 108
• Main Authors: Boudourides, Moses, Stevens, Andrew, Tsakonas, Giannis, Lenis, Sergios
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer International Publishing AG 2022; Springer International Publishing
• Series: Lecture Notes in Social Networks
• Subjects: Assortativity; Citation graphs; DAG; Nodal degrees; Scientometric networks; Topic modeling; Triadic closure and openness
• ISBN: 3031078683; 9783031078682
• ISSN: 2190-5428; 2190-5436
• DOI: 10.1007/978-3-031-07869-9_5

This paper presents a Data-Network Science study on a dataset of publications archived in The Semantic Scholar Open Research Corpus (S2ORC) database and categorized under the area of “Disease Control through Social Network Surveillance,” an area abbreviated from now on as “DCSNS.” In particular, our dataset consists of 10,866 documents (which are articles and reviews), retrieved through a Boolean search, published in the period from 1983, the first year of cataloguing such publications in S2ORC, to 2020. Retrieving also the corpus of abstracts of these documents (publications) and applying the standard LDA Topic Modeling technique, we found an optimal number of six topics producing the maximum topic coherence score among the corresponding topic models with varying numbers of topics. In that matter, the network of our study becomes a directed citation graph of publications in the area of DCSNS, with nodes/publications labeled by the Topics (into which Topic Modeling categorizes words from their abstracts). Our aim is to study global and local network properties with regards to clustering under triadic relationships amongst connected nodes/publications, and with regards to the assortativity of attributes related to the content of publications (such as types of publications and themes in the employed keyword searches). Thus, we have succeeded in analyzing the interplay between semantics and structure in the area of publications on DCSNS, by examining and discovering the occurrence of certain important attributes of publications in such a way that the aggregation of publications according to these attributes is associating the meaning of attribute affiliations to certain structural patterns of clustering, exhibited by the bibliographic citation network of the collected publications.