Infectious Diseases in Nigeria Using Topic Modelling: A Systematic Review
Cornelius Ayantse *
Computational Statistics Unit, Department of Statistics, University of Ibadan, Ibadan, Nigeria.
OlaOluwa S. Yaya
Computational Statistics Unit, Department of Statistics, University of Ibadan, Ibadan, Nigeria.
Aderonke Busayo Sakpere
Department of Computer Science, University of Ibadan, Nigeria.
Olubunmi K. D. Abel
Essex Business Intelligence Unit, NHS Arden and Greater East Midlands Commissioning Support Unit, United Kingdom.
IyanuOluwa O. Ojo
Faculty of Nursing, College of Medicine, University of Ibadan, Ibadan, Nigeria.
Okeke U. Joseph
Department of Mathematical Sciences, Taraba State University, Jalingo, Taraba State, Nigeria.
*Author to whom correspondence should be addressed.
Abstract
In this study, we used text mining to present a list of topics and diseases that have been most researched in the last two decades in the field of infectious diseases in Nigeria, using abstracts and keywords of publications. We accessed and analyzed 15,459 publications from the Scopus database, extracting out their abstracts, keywords, years of publication and titles. We used quanteda to get n-grams and document frequencies of the keywords, which produced 39 keywords for diseases based on unigram, 55 based on bigram, and 63 for trigram. HIV/AIDS and Malaria emerged as the two primary focus in this research. Taking advantage of technological advances, we used Chat GPT to check the prevalent infectious diseases in Nigeria and compared output with our result, and our search turned in 14 infectious diseases in Nigeria, out of which 10 were a match to the diseases contained in the research articles’ keywords. Thus, continuing measures to eradicate malaria in Nigeria, and interventions to reduce the prevalence of HIV/AIDS should be communicated. Using structural topic modelling, we were able to extract a total of 100 topics from the document abstracts in which topic 29 (sexual, HIV, partner, sex, condom, risk, men) stood out due to its higher distribution across the documents.
Keywords: Infectious disease, systematic literature review, topic modelling, text mining, Nigeria