|
Reference | Dataset | Techniques | Accuracy |
|
Moreno-Sanchez [7] | They used a dataset that comprises 299 patients who suffered heart failure. | Receiver operating characteristic (ROC) curve | 87.5% |
Vijayashree and Iyengar [18] | The UCI machine learning repository provided the HF Indian heart attack dataset. | Post hoc techniques | — |
Marimuthu et al. [19] | They used heart disease dataset. | -means and fuzzy -means clustering and random forest, XGBoost, and decision tree | 87% |
Golande and Kumar [20] | The data comes from a database of heart disease patients’ medical records. | SVM (support vector machine) | — |
Ayers et al. [21] | Training and validation datasets are used. | -nearest neighbors (KNN), naïve Bayes, and support vector machine (SVM) | 84.81% |
Haq et al. [22] | It is necessary to make use of a vast number of disparate electronic datasets. | Cardiac magnetic resonance (CMR) | — |
Mortazavi et al. [23] | | Logistic regression (LR) | 83.2%. |
Marbaniang et al. [24] | UCI provided a dataset on heart illness with 14 different features. | Random forest (RF), logistic regression, and support vector machine (SVM) | — |
Kathare and Gaikwad [13] | Heart study dataset was used. | Support vector machine (SVM) | 88.7% |
Segar et al. [3] | They gathered data from the website https://cran.r-project.org/web/packages/MASS/index.html. | Support vector machine (SVM) and -nearest neighbors (KNN) | 83.9% |
Chicco and Jurman [6] | Data from 299 people with heart failure is analysed by the researchers. | Stochastic gradient classifier | — |
Nashif et al. [25] | The dataset from the UCI machine learning repository was used. | Data mining modelling techniques | 85.1% and 79.5% |
Jindal et al. [26] | Using the UCI repository, a dataset with a patient’s medical history and attributes is picked. | Logistic regression, KNN, and random forest classifier | — |
Solanki and Sharma [27] | To get the data, they employed a variety of methods. | Artificial neural network (ANN) | 56.76% |
|