Key result
TANFIS classifier achieves ~99.8% accuracy for heart disease prediction, outperforming existing algorithms.
Why the study?
Feature extraction poses a major challenge for heart disease prediction because high-dimensional data increases learning times for existing machine learning classifiers.
Does a tuned adaptive neuro-fuzzy inference system (TANFIS) classifier improve the accuracy of heart disease prediction compared to existing machine learning algorithms?
Does a tuned adaptive neuro-fuzzy inference system (TANFIS) classifier improve the accuracy of heart disease prediction compared to existing machine learning algorithms?
The proposed TANFIS classifier demonstrates high accuracy (99.76%) for heart disease prediction on UCI repository datasets, outperforming existing machine learning algorithms.
Supports TANFIS development for heart disease prediction research; leaves open prospective clinical validation before any practice consideration.
In today's world, the advancement of telediagnostic equipment plays an essential role to monitor heart disease. The earlier diagnosis of heart disease proliferates the compatibility of treatment of patients and predominantly provides an expeditious diagnostic recommendation from clinical experts. However, the feature extraction is a major challenge for heart disease prediction where the high dimensional data increases the learning time for existing machine learning classifiers. In this article, a novel efficient Internet of Things‐based tuned adaptive neuro‐fuzzy inference system (TANFIS) classifier has been proposed for accurate prediction of heart disease. Here, the tuning parameters of the proposed TANFIS are optimized through Laplace Gaussian mutation‐based moth flame optimization and grasshopper optimization algorithm. The simulation scenario can be carried out using11 different datasets from the UCI repository. The proposed method obtains an accuracy of 99.76% for heart disease prediction and it has been improved upto 5.4% as compared with existing algorithms.
No takes yet. Share an insight, caveat, or question.
Sekar et al. (2021) studied Heart disease. Tuned adaptive neuro-fuzzy inference system (TANFIS) classifier vs. Existing algorithms was evaluated on Accuracy for heart disease prediction. The proposed TANFIS classifier achieved an accuracy of 99.76% for heart disease prediction, an improvement of up to 5.4% compared with existing algorithms.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: