Affordable Access

deepdyve-link
Publisher Website

Modeling Latent Topics in Social Media using Dynamic Exploratory Graph Analysis: The Case of the Right-wing and Left-wing Trolls in the 2016 US Elections.

Authors
  • Golino, Hudson1
  • Christensen, Alexander P2
  • Moulder, Robert3
  • Kim, Seohyun3
  • Boker, Steven M3
  • 1 University of Virginia, Charlottesville, USA. [email protected]
  • 2 University of Pennsylvania, Philadelphia, USA.
  • 3 University of Virginia, Charlottesville, USA.
Type
Published Article
Journal
Psychometrika
Publication Date
Mar 01, 2022
Volume
87
Issue
1
Pages
156–187
Identifiers
DOI: 10.1007/s11336-021-09820-y
PMID: 34757581
Source
Medline
Keywords
Language
English
License
Unknown

Abstract

The past few years were marked by increased online offensive strategies perpetrated by state and non-state actors to promote their political agenda, sow discord, and question the legitimacy of democratic institutions in the US and Western Europe. In 2016, the US congress identified a list of Russian state-sponsored Twitter accounts that were used to try to divide voters on a wide range of issues. Previous research used latent Dirichlet allocation (LDA) to estimate latent topics in data extracted from these accounts. However, LDA has characteristics that may limit the effectiveness of its use on data from social media: The number of latent topics must be specified by the user, interpretability of the topics can be difficult to achieve, and it does not model short-term temporal dynamics. In the current paper, we propose a new method to estimate latent topics in texts from social media termed Dynamic Exploratory Graph Analysis (DynEGA). In a Monte Carlo simulation, we compared the ability of DynEGA and LDA to estimate the number of simulated latent topics. The results show that DynEGA is substantially more accurate than several different LDA algorithms when estimating the number of simulated topics. In an applied example, we performed DynEGA on a large dataset with Twitter posts from state-sponsored right- and left-wing trolls during the 2016 US presidential election. DynEGA revealed topics that were pertinent to several consequential events in the election cycle, demonstrating the coordinated effort of trolls capitalizing on current events in the USA. This example demonstrates the potential power of our approach for revealing temporally relevant information from qualitative text data. © 2021. The Author(s).

Report this publication

Statistics

Seen <100 times