Business Intelligence Approach-Based Hybrid Deep Learning Model for QS World University Ranking

Asaad R.  Kareem; Hasanen S.  Abdullah

doi:10.24996/ijs.2025.66.1.30

Authors

Asaad R. Kareem Computer Science Department, the University of Technology, Baghdad, Iraq
Hasanen S. Abdullah Computer Science Department, the University of Technology, Baghdad, Iraq

DOI:

https://doi.org/10.24996/ijs.2025.66.1.30

Keywords:

business intelligence BI, Convolutional Neural Networks CNN, Long Short-Term Memory LSTM, Hybrid deep learning

Abstract

The QS World University Rankings have become the most widely used basis for comparing universities across the world. Selecting a university is a difficult decision, and this may be particularly true for international students. The rankings of universities have been designed to help students make more informed choices. Informed decision institutions face challenges in analyzing huge data volumes and ensuring generalizability. This study presents a hybrid deep learning model (CNN-LSTM) that addresses these issues, minimizing data requirements and ensuring compatibility. The proposed hybrid deep learning model was compared with five other machine learning models. It demonstrated the highest efficiency, with a mean squared error of 0.0144 and an accuracy of 98.5%.