Abstract
Heart disease remains a significant challenge in the medical field, particularly in predictive diagnostics. This research aims to present a comprehensive investigation into the development and evaluation of a novel approach for heart disease detection using a Weighted k-Nearest Neighbors (WKNN) method. The method employs Euclidean distance metrics and Gaussian kernel weighting for optimal classification results. The research dataset consists of 200 data points, each with 10 key indicators such as age, sex, chest pain type, resting blood pressure, cholesterol levels, fasting blood sugar, resting electrocardiographic results, maximum heart rate achieved, exercise-induced angina, and ST depression relative to rest. Through rigorous experimentation, it is identified that the optimal value of K for classification is 11, with a sigma value of 1.5 for the Gaussian kernel weighting. During the training and evaluation phase, the proposed WKNN method achieved impressive performance metrics, with an accuracy of 91.8%, precision of 93%, and recall of 91%. These findings underscore the potential of the WKNN model as a reliable tool for heart disease detection, showing great promise for practical application in clinical settings. The results emphasize that the proposed method can contribute significantly to improving diagnostic accuracy for heart disease patients
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call