Dataset for Semester-wise GPA Prediction Using Academic, Behavioral, and Socio-Demographic Features (CSE Students, Dhaka International University)

This dataset contains undergraduate student records collected from the Department of Computer Science and Engineering (CSE) at Dhaka International University. It is designed to support research in educational data mining and machine learning for academic performance prediction. The dataset consists of 414 student instances and 35 features, including academic history, behavioral attributes, and socio-demographic characteristics. The target variable is the current semester Grade Point Average (GPA), making the dataset suitable for regression-based predictive modeling tasks. Categorical variables have been preprocessed using label encoding, and the dataset is provided in a structured tabular format. It can be used to develop and evaluate machine learning models as well as for explainability analysis using techniques like SHAP (SHapley Additive exPlanations). This dataset enables research on early identification of academically at-risk students, supporting data-driven intervention strategies in higher education.All data have been anonymized to ensure privacy and confidentiality, and no personally identifiable information (PII) is included.