Table 7.
Accuracy comparison of our proposed breast cancer prediction models with previous studies that used the same WDBC dataset.
| Author Name | Reference | Year | Model/Method | Best Observed Accuracy |
|---|---|---|---|---|
| Maglogiannis I. et al. | [43] | 2007 | SVM Gaussian RBF | 97.54% |
| Mert et al. | [15] | 2015 | KNN | 92.56% |
| Hazra et al. | [29] | 2016 | Support Vector Machine (using 19 features) | 94.423% |
| Osman A. H. et al. | [44] | 2017 | SVM | 95.23% |
| Wang et al. | [30] | 2018 | SVM based ensemble learning | 96.67% |
| Abdar et al. | [16] | 2018 | Nested Ensemble 2-MetaClassifier (K = 5) | 97.01% |
| Mushtaq et al. | [18] | 2019 | KNN with multiple distances (Correlation K = 2) | 91.00% |
| Rajaguru & Chakravarthy | [17] | 2019 | KNN Euclidean distance | 95.61% |
| Durgalakshmi & Vijayakumar | [28] | 2019 | SVM | 73% |
| Khan et al. | [45] | 2020 | SVM | 97.06% |
| Al-Azzam & Shatnawi | [34] | 2021 | LR with area under curve | 96% |
| Proposed Prediction Models | 2022 | Polynomial SVM | 99.03% | |
| LR with RFE | 98.06% | |||
| Voting Classifier (CV) | 97.61% | |||
| KNN Performance with hyperparameter | 97.35% | |||
* The bold number indicate the top performance of the classifiers.