Article Details
  • Published Online:
    June  2025
  • Product Name:
    The IUP Journal of Information Technology
  • Product Type:
    Article
  • Product Code:
    IJIT010625
  • DOI:
    10.71329/IUPJIT/2025.21.2.7-23
  • Author Name:
    Chinelo Blessing Oribhabor and Celestine Uche Agwi
  • Availability:
    YES
  • Subject/Domain:
    Engineering
  • Download Format:
    PDF
  • Pages:
    7-23
Volume 21, Issue 2, April-June 2025
Application of ML Techniques in Detecting Reference Virus Phenomenon: An Academic Review
Abstract

The paper evaluates and synthesizes current academic research on the application of machine learning (ML) techniques in detecting and mitigating the reference virus. It examines various literature on how ML methods, specifically natural language processing (NLP) models, are used to identify and address citation problems in academic papers. The findings underscore that no single ML paradigm is sufficient on its own. ML methods, particularly deep learning (DL) algorithms based on graphs and models based on NLP, can automatize the identification of citation errors and decrease the dissemination of incorrect references. Nevertheless, obstacles like data accessibility, computational intricacy, and comprehensibility of ML models endure. This paper offers important insights into the upcoming trends in ML-based citation management and emphasizes the necessity of combining technology with traditional scholarly methods to protect academic honesty. The paper concludes that ML offers a promising and scalable solution for detecting the reference virus phenomenon in academic publications, helping to enhance citation integrity and uphold scholarly standards.

Introduction

In the evolving landscape of academic publishing and digital scholarship, one troubling phenomenon that has emerged is the “reference virus”. This term refers to the proliferation and perpetuation of inaccurate, irrelevant, or nonexistent citations in scholarly literature.