Table 1 Major findings of the study.
From: Revolutionizing heart disease prediction with quantum-enhanced machine learning
S. no. | References | Algorithms | Major findings |
|---|---|---|---|
1 | SVM | SVM is most preferred as it can handle high dimensionality problems Boosting SVM enhances accuracy up to 99.75% | |
2 | NBC | NBC's heart disease prediction is challenging. Since requires all features to be mutually independent | |
3 | ANN | ANN is good at generalization and capable of analyzing complex data and able to attain an accuracy of up to 95.82% To eliminate the block box nature of ANN, feature selection is employed which enhances prediction accuracy further | |
4 | KNN | KNN is simple and not suitable for high dimensional data which is common in heart disease datasets In high-dimensional data, the distance between points tends to become less informative, making it difficult for KNN to identify the nearest neighbors accurately | |
5 | DT | It can be prone to overfitting and it is inappropriate for handling data with missing values DT can sometimes be biased toward selecting features resulting in a less accurate model | |
6 | LR | It requires a linear relationship between features It can limit the capability to capture underlying patterns leading to lower accuracy | |
7 | RF | The Potential issue with RF is that it can be computationally expensive and memory-intensive, particularly when dealing with medical data | |
8 | Ensemble Approaches | The Bagging Ensemble approach yielded greater accuracy up to 97.57% due to its ability to reduce overfitting, improve generalization, and combine the strengths of multiple machine learning algorithms | |
9 | Quantum Computing | It performs certain operations faster This speedup can be harnessed to accelerate certain machine-learning algorithms by reducing the time required for training |
