Institution profile

Florida Atlantic University

Academic institutionnorthamerica · us
Official website
Research library52linked papers
Opportunities0open roles
Selected work

Representative Papers

Predicting Anemia Among Under-Five Children in Nepal Using Machine Learning and Deep Learning

Feb 01, 2026

This study addresses the high prevalence of anemia among children under five in Nepal by leveraging the 2022 Nepal Demographic and Health Survey data. To balance model interpretability with the identification of key risk factors, a consensus feature set was constructed through the integration of four feature selection methods: chi-square test, mutual information, point-biserial correlation, and Boruta. The binary classification performance of logistic regression, XGBoost, support vector machines (SVM), deep neural networks (DNN), and TabNet was systematically evaluated on imbalanced data. Results indicate that logistic regression achieved the highest F1-score (0.649) and recall (0.701), SVM attained the best AUC (0.736), and DNN yielded the highest accuracy (0.709), collectively demonstrating the feasibility and potential of machine learning approaches for early screening of childhood anemia.

1 citationsRead paper

Predicting Student Success with Heterogeneous Graph Deep Learning and Machine Learning Models

Jan 11, 2026arXiv.org

This study addresses the challenge of enabling early and continuous prediction of academic success by leveraging multi-source, heterogeneous student data that includes dynamic assessment features and diverse entity relationships. To this end, the authors propose a novel framework that integrates heterogeneous graph neural networks with traditional machine learning methods, uniquely incorporating dynamic assessment features into heterogeneous graph modeling. Complex dependencies among students, courses, and assessments are captured through carefully designed meta-paths. Evaluated on the OULA dataset, the approach achieves a validation F1-score of 68.6% using only 7% of semester data and reaches 89.5% by the semester’s end, outperforming the best conventional model by 4.7% in early prediction performance. These results demonstrate the method’s effectiveness and innovation in modeling temporal and relational student data for academic outcome forecasting.

1 citationsRead paper
Recent publications

Latest Papers