Conference Paper2023Cited by 16

SMOTE Oversampling and near Miss Undersampling Based Diabetes Diagnosis from Imbalanced Dataset with XAI Visualization

Nasim Mahmud Nayan, Ashraful Islam, Muhammad Usama Islam, Eshtiak Ahmed, Mohammad Mobarak Hossain, Md Zahangir Alam

28th IEEE Symposium on Computers and Communications (ISCC)-ICT4eHealth

IEEE, pp. 1-6

  • Good Health and Well-being — Relevance: 95%
DOI: 10.1109/iscc58397.2023.10218281URL

Abstract

This study investigated the predictive ability of ten different machine learning (ML) models for diabetes using a dataset that was not evenly distributed. Additionally, the study evaluated the effectiveness of two oversampling and undersampling methods, namely the Synthetic Minority Oversampling Technique (SMOTE) and the Near-Miss algorithm. Explainable Artificial Intelligence (XAI) techniques were employed to enhance the interpretability of the model's predictions. The results indicate that the extreme gradient boosting (XGB) model combined with SMOTE oversampling technique exhibited the highest accuracy and an F1-score of 99% and 1.00 respectively. Furthermore, the utilization of XAI methods increased the dependability of the model's decision-making process, rendering it more appropriate for clinical use. These results imply that integrating XAI with ML and oversampling techniques can enhance the early detection and management of diabetes, leading to better diagnosis and intervention.

Keywords

  • Undersampling
  • diabetes diagnosis
  • Explainable AI
  • Oversampling
  • XGBoost
  • early detection of diabetes
  • SMOTE
  • class imbalance
  • model interpretability
  • near misses

CCDS Authors

Ashraful Islam, PhD
Ashraful Islam

Ashraful Islam, PhD

  • Center Director, CCDS
  • Director, HCI Wing

Assistant Professor

Department of Computer Science and Engineering

Independent University, Bangladesh (IUB)

Published at

CHI, UIST, UbiComp, IJCNN, ICIP, HCII, ICDMW, BLP, IEEE Internet of Things Journal, JMIR, JMIR Human Factors, JMIR Formative Research, IEEE Access, Smart Agricultural Technology, Diabetes Research and Clinical Practice, PLOS ONE

Research Interests

Human-Computer Interaction, AI for Social Good, AI for Public Health, AI for Impact

Muhammad Usama Islam, PhD
Muhammad Usama Islam

Muhammad Usama Islam, PhD

  • Collaborator, HCI Wing

Assistant Professor

+1 more affiliation

Metro State University

Eshtiak Ahmed
Eshtiak Ahmed

Eshtiak Ahmed

  • Collaborator, HCI Wing

Doctoral Researcher

Gamification Group

Tampere University, Finland

Published at

UbiComp, ICIP, HCII, IEEE Access

Research Interests

Human-Computer Interaction, Human-Robot Interaction, Gamification and Assistive Technologies

References

  1. 1.Nitesh V. Chawla, Kevin W. Bowyer, Lawrence Hall, W. Philip Kegelmeyer. (2002). SMOTE: Synthetic Minority Over-sampling Technique. Journal of Artificial Intelligence Research, 16, 321–357[10.1613/jair.953]
  2. 2.Girish Chandrashekar, Ferat Sahin. (2013). A survey on feature selection methods. Computers & Electrical Engineering, 40(1), 16–28[10.1016/j.compeleceng.2013.11.024]
  3. 3.J. Pradeep Kandhasamy, S. Balamurali. (2015). Performance Analysis of Classifier Models to Predict Diabetes Mellitus. Procedia Computer Science, 47, 45–51[10.1016/j.procs.2015.03.182]
  4. 4.Ronald J. Tallarida, Rodney B. Murray. (1987). Chi-Square Test. , 140–142[10.1007/978-1-4612-4974-0_43]
  5. 5.Show-Jane Yen, Yue‐Shi Lee. (2006). Under-Sampling Approaches for Improving Prediction of the Minority Class in an Imbalanced Dataset. Lecture notes in control and information sciences, 731–740[10.1007/978-3-540-37256-1_89]
  6. 6.N. Yuvaraj, K R SriPreethaa. (2017). Diabetes prediction in healthcare systems using machine learning algorithms on Hadoop cluster. Cluster Computing, 22(S1), 1–9[10.1007/s10586-017-1532-x]
  7. 7.Luis Fregoso-Aparicio, Julieta Noguez, Luis Montesinos, José Antonio García-García. (2021). Machine learning and deep learning predictive models for type 2 diabetes: a systematic review. Diabetology & Metabolic Syndrome, 13(1), 148[10.1186/s13098-021-00767-9]
  8. 8.S K Somasundaram, P. Alli. (2017). A Machine Learning Ensemble Classifier for Early Prediction of Diabetic Retinopathy. Journal of Medical Systems, 41(12), 201[10.1007/s10916-017-0853-x]
  9. 9.Pratya Nuankaew, Supansa Chaising, Punnarumol Temdee. (2021). Average Weighted Objective Distance-Based Method for Type 2 Diabetes Prediction. IEEE Access, 9, 137015–137028[10.1109/access.2021.3117269]
  10. 10.Manju Devi. (2016). ANALYSIS OF VARIOUS DATA MINING TECHNIQUES TO PREDICT DIABETES MELLITUS. , 11(1), 727–730
  11. 11.Muhammad Usama Islam, Md Mozaharul Mottalib, Mehedi Hassan, Zubair Ibne Alam, SM Zobaed, Md. Fazle Rabby. (2022). The Past, Present, and Prospective Future of XAI: A Comprehensive Review. Studies in computational intelligence, 1–29[10.1007/978-3-030-96630-0_1]

Cited By

  1. 1.Aymin Javed, Nadeem Javaid, Muhammad Hasnain, Umair Sarfraz, Imran Ahmed, Muhammad Shafiq, Jin-Ghoo Choi. (2024). Applying Advanced Data Analytics on Pregnancy Complications to Predict Miscarriage With eXplainable AI. IEEE Access, 12, 176845–176862[10.1109/access.2024.3486058]
  2. 2.Nasim Mahmud Nayan, Al Mamun Rana, Md. Monirul Islam, Jia Uddin, Jia Uddin, Tahmina Yasmin, Jasim Uddin, Jasim Uddin. (2025). An interpretable and balanced machine learning framework for Parkinson’s disease prediction using feature engineering and explainable AI. PLoS ONE, 20(10), e0333418[10.1371/journal.pone.0333418]
  3. 3.Mohammad Mobarak Hosaain, Mohammod Abdul Kashem, Nasim Mahmud Nayan. (2024). Artificial Intelligence-Driven Approach for Predicting Maternal Health Risk Factors. , 153–158[10.1109/seeda-cecnsm63478.2024.00035]
  4. 4.Mark Lokanan. (2024). Exploring Resampling Techniques in Credit Card Default Prediction. Research Square[10.21203/rs.3.rs-4087259/v1]
  5. 5.Rupali Ahuja, Gaurav Indra. (2024). Leveraging XAI for Discovering Crucial Demographic, Clinical and Pathological Diabetes Mellitus Biomarkers. , 1–6[10.1109/delcon64804.2024.10866967]
  6. 6.Rizwan Alam, Mohammad Atif, Naeem Ahmad, Tejas V. Kalal, Amisha Khune. (2025). Enhancing early-stage diabetes prediction with Explainable Artificial Intelligence(XAI). , 1–6[10.1109/icaihc64101.2025.10957089]
  7. 7.Mark Lokanan. (2025). Introducing DART: A Novel Deep Adaptive Upsampling Technique for Handling Class Imbalance. Intelligent systems in accounting, finance and management/Intelligent systems in accounting, finance & management, 32(3)[10.1002/isaf.70014]
  8. 8.Mohammad Mobarak Hossain, Nasim Mahmud Nayan, Mohammod Abdul Kashem. (2026). A comprehensive maternal health risk prediction dataset from IoT-enabled medical cyber-physical systems in developing countries: supporting machine learning and deep learning applications for clinical decision support. BMC Medical Informatics and Decision Making, 26(1)[10.1186/s12911-026-03343-1]
  9. 9.Andleeb Khan, Saundarya Yadav, Parma Nand, Akib Mohi Ud Din Khanday, Bharat Bhushan, Akhtar Jamil, Alaa Ali Hameedkhan. (2024). An Explainable Predictive Model for Diabetes Detection Using Shapley Additive Explanations Approach. Lecture notes in networks and systems, 304–317[10.1007/978-3-031-70924-1_23]
  10. 10.Sena Çeper Özçelik, Gülay Tezel. (2026). Analysis of the effects of batch normalization and the NearMiss algorithm on sleep stage classification. Signal Image and Video Processing, 20(9)[10.1007/s11760-026-05573-2]
  11. 11.Faiza Nawaz, Hamna Mughal, Arbab Fatima. (2026). Prevalence-Aware Machine Learning for Risk Assessment of Rare Medical Conditions. , 1–6[10.1109/icomet69771.2026.11591774]
  12. 12.Lubana Al Rayes, Menatalla Haggag, Imad Afyouni. (2024). Predicting Pre-Diabetic and Diabetes in Adults and Elderlies Using Machine Learning. , 1–8[10.1109/aset60340.2024.10708714]
  13. 13.Hezam Albaqami, Muhammad Asif Ayub, Nasir Ahmad, Yaseen Ahmad, Mohammad M. Alqahtani, Abdullah M. Algamdi, Almoaid A. Owaidah, Kashif Ahmad. (2026). Stylometry Analyzis of Human and Machine Text for Academic Integrity. Computers, 15(4), 217[10.3390/computers15040217]
  14. 14.Flavia Maria G. S. A. Oliveira, Sandro Muniz Cavalcanti, Michael C. K. Khoo. (2025). Cardiorespiratory Markers for Early Detection of Type 2 Diabetes: Machine Learning Models (Preprint). [10.2196/preprints.82084]
  15. 15.Flavia Maria G S A Oliveira, Sandro Muniz Cavalcanti, Michael C K Khoo. (2025). Cardiorespiratory Markers of Type 2 Diabetes: Machine Learning–Based Analysis. JMIR Diabetes, 11, e82084[10.2196/82084]
  16. 16.Mark Lokanan. (2025). Introducing DART: A Novel Deep Adaptive Upsampling Technique for Handling Class Imbalance. Research Square[10.21203/rs.3.rs-6895500/v1]
← All publications