Comparing the Behaviour of Two Topic-Modelling Algorithms in COVID-19 Vaccination Tweets: LDA vs. LSA

Comparing the Behaviour of Two Topic-Modelling Algorithms in COVID-19 Vaccination Tweets: LDA vs. LSA

Jordan Thomas Bignell, Georgios Chantziplakis, Alireza Daneshkhah
Copyright: © 2022 |Pages: 20
DOI: 10.4018/IJoSE.292445
OnDemand:
(Individual Articles)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Coronavirus is a newly developed infectious disease that has triggered a pandemic due to its ease of transmission as of early 2020. Several groups from various countries have been working on a vaccine to prevent and avoid the spread of the virus in this outbreak. In this article, the main objective is to compare LDA against LSA to gain a better understanding of the Tweets and which Topic Modelling technique fits best for this task, additionally if the feedback of the Tweets were positive or negative sentiment. It was concluded that LDA was a better-unsupervised technique for categorizing the raw text in 12 topics.
Article Preview
Top

Literature Review

In this publication by Garcia & Berton (2021), the paper was based on tweets in both English and Portuguese language, which from research studies have only been conducted in one language. The modelling techniques used in this paper were Latent Dirichlet Allocation (LDA) and Gibbs Sampling Dirichlet Multinomial Mixture (GSDMM) which is better suited for short text due to the lack of word co-occurrences.

Complete Article List

Search this Journal:
Reset
Volume 6: 1 Issue (2024): Forthcoming, Available for Pre-Order
Volume 5: 2 Issues (2022): 1 Released, 1 Forthcoming
Volume 4: 2 Issues (2021)
Volume 3: 2 Issues (2020)
Volume 2: 2 Issues (2019)
Volume 1: 2 Issues (2018)
View Complete Journal Contents Listing