Heart disease prediction using KNN algorithm approach
|
Full Text |
Pdf
|
|
Author |
S. Ghouhar Taj and K. Kalaivani
|
|
e-ISSN |
1819-6608 |
|
On Pages
|
754-764
|
|
Volume No. |
19
|
|
Issue No. |
12
|
|
Issue Date |
September 3, 2024
|
|
DOI |
https://doi.org/10.59018/062498
|
|
Keywords |
MLP, KNN, choice tree, random forest, cardiac database.
|
Abstract
Machine getting to know and records-based totally techniques for predicting and diagnosing coronary heart ailment may be an extraordinary medical gain, but it's far a chief undertaking to improve. In many countries there may be a shortage of cardiovascular professionals and a giant quantity of instances of misdiagnosis may be addressed by way of setting up correct and powerful early-stage cardiac forecasts with the aid of medical decision-making analysis of virtual patient information. This has a look at aimed picking out excessive-overall performance devices getting to know variants for such diagnostic purposes. Several gadget-getting-to-know algorithms have been used which might be in comparison and done with accuracy and accuracy in predicting heart disease. The scores of the importance of each detail are restrained to all algorithms used except MLP and KNN. All elements are calculated based totally on the cost points to find those who provide high-danger coronary heart disease prognosis. The look discovered that using Kaggle's 3-segment cardiac database based on pro-k (KNN), choice tree (DT), and random Forests (RF) RF technique algorithms done ninety-seven.2% accuracy and 97.2% sensitivity to make clear. Therefore, we've observed that an easily supervised machine gaining knowledge of a set of rules can be wielded to make coronary heart disease conjecture with the very best accuracy and the most satisfactory possible use.
Back