Heart Disease Prediction Using Weighted K-Nearest Neighbor Algorithm

Khalidou Abdoulaye Barry (), Youness Manzali (), Mohamed Lamrini (), Flouchi Rachid () and Mohamed Elfar ()
Additional contact information
Khalidou Abdoulaye Barry: USMBA
Youness Manzali: USMBA
Mohamed Lamrini: USMBA
Flouchi Rachid: USMBA
Mohamed Elfar: USMBA

SN Operations Research Forum, 2024, vol. 5, issue 3, 1-16

Abstract: Abstract Heart disease includes many kinds of conditions affecting the heart and has been the main cause of death worldwide in recent decades. To prevent further damage and preserve patients’ lives, it is crucial to detect heart disease early and adequately. Consequently, we require an approach capable of predicting heart disease before it progresses to a critical stage. Researchers in the field of medical sciences are interested in machine learning. They use several machine learning algorithms and methodologies for predicting heart disease. A weighted K-nearest neighbor model using feature scores is proposed in this study to increase classification accuracy. Initially, we used the K-means clustering method to group similar features, selecting the one with the highest relevance in each group. Following this selection, we employed the relief feature selection technique to determine the scores, which will be used as weights. The proposed approach is applied to five aggregated datasets from IEEE Data Port or Kaggle, including Hungarian, Cleveland, Long Beach VA, Switzerland, and Statlog (Heart). Several performance metrics, such as the AUC curve, recall, precision, F1-score, and accuracy, are utilized to assess the efficacy and strength of the constructed model. The underlying study concluded that the weighted KNN has outperformed all other algorithms used in this study for comparison across all measures. It had the best accuracy of 93.28%.

Keywords: Heart disease; Machine learning; Clustering; Feature selection (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s43069-024-00356-2 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:snopef:v:5:y:2024:i:3:d:10.1007_s43069-024-00356-2

Ordering information: This journal article can be ordered from
https://www.springer.com/journal/43069

DOI: 10.1007/s43069-024-00356-2

Access Statistics for this article

SN Operations Research Forum is currently edited by Marco Lübbecke

More articles in SN Operations Research Forum from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().