Predicting Breast Cancer Leveraging Supervised Machine Learning Techniques

Table 7

Comparison of machine learning algorithms on Wisconsin Breast Cancer dataset.

Author	Year	Dataset	Imbalance handling	Feature selection	Features	Classifier	Validation type	Accuracy achieved %
Wisconsin original breast cancer dataset (WOBC) [23]
Salama et al. [41]	2012	WOBC	—	Chi-square & PCA	10	J48 & MLP	10-fold	97.28%
Hamsagayathri & Sampath [42]	2017	WOBC	—	Feature ranking	—	Random Forest	10-fold	96.70%
Our approach	2020	WOBC	Normalization by standardization	Correlation based selection & RFE	8	MLP	5-fold	98.20%
Wisconsin Prognostic breast cancer data (WPBC) [24]
Tintu and Paulin [43]	2013	WPBC	Manual removal of instances	Feature ranking	—	Fuzzy -means clustering	4-fold	97.13%
Khan et al. [44]	2013	WPBC	—	YAGGA	19	Linear regression	10-fold	84.34%
Our approach	2020	WPBC	Normalization by standardization	Correlation-based selection and RFE	16	MLP	5-fold	98.33%