Machine Learning-Based Tertiary Institutions’ Data Harmonization with a Large Language Model Decision Support System: A Literature Review

Abstract

The increasing complexity of data ecosystems in tertiary institutions necessitates advanced approaches to data harmonisation. This literature appraisal investigates the purposes of Machine Learning (ML) and Large Language Models (LLMs) as groundbreaking assets for enabling data-oriented decision-making and integration. It synthesizes current research on their applications in data cleaning, feature selection, predictive analytics, and the processing of unstructured data. The findings indicate that ML and LLMs improve data quality, interoperability, and institutional responsiveness. Still, issues of morality like the bias present in algorithms, apprehensions regarding data privacy, and the imperative for clarity persist as major concerns. The review concludes with recommendations for future research to guide the responsible and effective deployment of ML and LLMs in higher education data harmonization frameworks.