Degree
Master of Science in Data Science
Department
Department of Computer Science
Faculty/ School
School of Mathematics and Computer Science (SMCS)
Date of Submission
Spring 2024
Supervisor
Dr. Muhammad Sarim, Visiting Faculty, Department of Computer Science, School of Mathematics and Computer Science (SMCS)
Keywords
Data Science, Predictive Modeling, Machine Learning, Natural Language Processing, Sentiment Analysis, Admissions Process, Higher Education
Abstract
This study aims to enhance the efficiency and effectiveness of the admissions process at the Institute of Business Administration (IBA) with the application of data science techniques through the development of a predictive modeling framework. The study focuses on a primary objective: forecasting the likelihood of an undergraduate applicant's success or failure in the interview process at IBA. The methodology involves the collection and preprocessing of a rich and diverse dataset of undergraduate applicant information, including demographic profiles, academic grades, socio-economic backgrounds, and personal statements. The dataset is then subjected to rigorous cleaning and exploratory data analysis (EDA) to gain a deeper understanding of the underlying patterns. A series of predictive models are developed and evaluated using boosting machine learning algorithms, including Histogram Gradient Boosting, XGBoost and CatBoost. The models are optimized through hyperparameter tuning and crossvalidation to ensure their robustness, generalizability, and predictive accuracy. In addition to traditional data analysis techniques, the study incorporates sentiment analysis by fine tuning transformer-based natural language processing (NLP) models: DistilBERT and RoBERTa, to extract meaningful insights from the textual data in personal statements. The study's findings are expected to provide valuable insights and actionable recommendations for the admissions committee at IBA, enabling them to make more informed, data-driven, and unbiased decisions when it comes to accepting undergraduate applicants. Furthermore, the research contributes to the broader literature on the applications of data science, sentiment analysis and predictive modeling in the field of higher education, particularly in the context of admissions.
Document Type
Restricted Access
Submission Type
Research Project
Recommended Citation
Ahmad, A. (2024). Students’ Admission Forecasting: Analyzing Admission Trends and Predicting University Acceptance (Unpublished graduate research project). Institute of Business Administration, Pakistan. Retrieved from https://ir.iba.edu.pk/research-projects-msds/32
Loading...
The full text of this document is only accessible to authorized users.
