Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
10.55041/IJSREM.NCET016
Volume:
10 |
10 |
Oct 2026
Pages:
1-7
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Computer Science & Engineering
Syed Ammal Arts and Science College
Computer Science & Engineering
Syed Ammal Arts and Science College
Computer Science & Engineering
Syed Ammal Arts and Science College
Heart disease prediction from routinely collected clinical variables is an important research problem because machine learning can assist clinicians in identifying patients who may require closer assessment. However, predictive performance can be affected by missing values, redundant attributes, class distribution, feature scaling, and the limited interpretability of complex models. This paper proposes a reproducible hybrid machine learning framework that combines data preprocessing, two-stage feature selection, comparative classification, soft-voting model fusion, and explainable artificial intelligence. The framework is designed for binary heart-disease prediction using the Cleveland subset of the UCI Heart Disease dataset. The candidate classifiers are Logistic Regression, Support Vector Machine, Random Forest, and Gradient Boosting. Mutual-information filtering is used as an initial relevance screen, followed by recursive feature elimination to obtain a compact predictive feature set. The selected features are supplied to the individual classifiers and to a probability-based soft-voting ensemble. Model performance is assessed using accuracy, precision, recall, F1-score, ROC-AUC, and confusion-matrix analysis. SHAP-based explanations are incorporated to estimate global feature importance and patient-level contribution patterns. Unlike a paper that reports unverified or fabricated performance values, this manuscript specifies a complete experimental protocol and leaves the final numerical results to be generated from the stated dataset and implementation. The proposed framework is intended to provide a transparent and reproducible decision-support approach that can be independently validated before clinical use.
KEYWORDS
This paper presented a hybrid machine learning framework for early heart disease risk prediction based on preprocessing, two-stage feature selection, comparative classification, soft-voting model fusion, and SHAP-based explanation. The framework is designed to address three recurring issues in medical machine learning: noisy or redundant variables, difficulty in comparing heterogeneous classifiers under a common protocol, and limited interpretability of complex predictive systems. The proposed workflow is reproducible and explicitly avoids unsupported performance claims. The next required step is to execute the protocol on the UCI Cleveland dataset, report the resulting metrics with confidence intervals, generate the confusion matrix and ROC curve, and validate the selected model on an independent external dataset. Only after those steps should numerical performance claims be used in a publication submission.
[1] T. Banerjee and İ. Paçal, “A systematic review of machine learning in heart disease prediction,” Turkish Journal of Biology, vol. 49, no. 5, pp. 600–634, 2025, doi: 10.55730/1300-0152.2766. [2] X. Dai et al., “Machine learning approaches for cardiovascular disease prediction: A review,” Archives of Cardiovascular Diseases, 2025, doi: 10.1016/j.acvd.2025.04.055. [3] C. Cortes and V. Vapnik, “Support-vector networks,” Machine Learning, vol. 20, pp. 273–297, 1995, doi: 10.1007/BF00994018. [4] L. Breiman, “Random forests,” Machine Learning, vol. 45, pp. 5–32, 2001, doi: 10.1023/A:1010933404324. [5] J. H. Friedman, “Greedy function approximation: A gradient boosting machine,” Annals of Statistics, vol. 29, no. 5, pp. 1189–1232, 2001, doi: 10.1214/aos/1013203451. [6] S. M. Lundberg and S.-I. Lee, “A unified approach to interpreting model predictions,” in Advances in Neural Information Processing Systems 30, 2017, pp. 4765–4774. [7] A. Janosi, W. Steinbrunn, M. Pfisterer, and R. Detrano, “Heart Disease,” UCI Machine Learning Repository, 1989/updated repository record, doi: 10.24432/C52P4X. [8] scikit-learn developers, “Metrics and model evaluation documentation,” scikit-learn, accessed 2026.
Publication Date
10/5/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
1
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...