Adaptive Sequence Pattern Based Hyperfeature Evaluation Framework for Heart Disease Risk Prediction in Diabetic Patients

Authors

  • K.Sathiesh Kumar Author
  • Dr.Geetha. K Author

Keywords:

Diabetes, Feature Selection, Heart Disease Prediction, Machine Learning, Preprocessing

Abstract

Heart disease is considered to be one of the main causes of death among diabetic patients and, therefore, early identification and evaluation of risks are essential in case of successful clinical intervention. The research proposes a novel structure of heart disease prediction through the datasets related to diabetes through the combination of the advanced preprocessing and feature selection techniques. First, an Adaptive Sequence Pattern Preprocessor (A-SPP) is used to improve the quality of data by means of deleting duplicate records, managing missing values, outliers, and transforming categorical attributes into numerical ones. Moreover, normalization is used to have equal-valued features and to enhance learning of models. The HyperFeature Evaluation Framework (HFEF) is then presented following preprocessing where the most informative and stable clinical attributes are determined on the basis of predictive contribution, stability, and redundancy analysis. Such a strategy helps to decrease the data dimensions and maintain medically significant features. Empirical testing of the framework of two benchmark datasets show that the proposed framework has a significant positive impact on the prediction performance with the accuracy scores of 0.939 and 0.943, respectively. In general, the proposed framework offers an effective and stable solution to the risk prediction of early heart disease in diabetic patients, which could contribute to better clinical practices.

Downloads

Download data is not yet available.

Downloads

Published

2026-09-22

How to Cite

Adaptive Sequence Pattern Based Hyperfeature Evaluation Framework for Heart Disease Risk Prediction in Diabetic Patients . (2026). Minimax Theory and Its Applications, 11(2s), 200-220. https://journalmta.com/index.php/jmta/article/view/409