An explainable machine learning approach for automated medical decision support of heart disease
This paper introduces an interpretable machine learning model using SHapley Additive exPlanations (SHAP) for predicting heart disease, comparing various classifiers and parameter tuning techniques. The model matches existing performance while ensuring transparency and reproducibility for future research.
Published on Data & Knowledge Engineering Journal on 8 July 2024
Machine learning techniques to predict the risk of developing diabetic nephropathy: a literature review
Conducted research on Machine Learning approaches to enhance Diabetic Nephropathy risk prediction through comprehensive temporal analysis of clinical data, emphasizing the crucial integration of Electronic Health Records for more reliable models.
Published on Journal of Diabetes & Metabolic Disorders on 5 December 2023
Depression Detection Using Deep Learning and Natural Language Processing Techniques: A Comparative Study
This research suggests the use of Natural Language Processing (NLP) techniques to spot signs of depression in tweets, finding success with an 84.83% accuracy using a method called Extra Trees combined with TF-IDF feature extraction method.
Presented at CIARP 2023: Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications on 29 November 2023
Oversampling Techniques for Diabetes Classification: a Comparative Study
A comparative study on the combination of SMOTE oversampling technique variants and machine learning algorithms for diabetes prediction using the unbalanced “PIMA Indian Diabetes” dataset.
Presented at the 2021 International Conference on e-Health and Bioengineering (EHB) on 31 December 2021.
Predicting Type 2 Diabetes Through Machine Learning: Performance Analysis in Balanced and Imbalanced Data
Addressing the prediction of Type 2 diabetes using the “PIMA Indians Diabetes” dataset. The study involves creating a balanced dataset and evaluating various machine learning methods.
Presented at the International Symposium on Ubiquitous Networking on 11 December 2021.