Article Preview
TopPrevious studies refer to a number of techniques to diagnose breast cancer pattern. Neural network, a classification method based on which many algorithms (Chou et al., 2004; Karabatak and Ince, 2009; Seral et al., 2007; Marcano-Cedeno et al., 2011; Abbass, 2002; Tuba and Yildirim, 2004, Shukla et al., 2018) have been developed for diagnosing breast cancer. Artificial neural networks and multivariate adaptive regression splines approach (Chou et al., 2004), Association rules and Neural network approach (Karabatak and Ince, 2009), radial basis function neural network classification technique (Subashini et al., 2009), Genetic algorithm based approach (Pena-Reyes and Sipper, 1999) and support vector machines (SVM) (Polat and Gunes, 2007; Akay, 2009; Majid et al., 2014; Maglogiannis et al., 2009; Zheng et al., 2014) are some of the techniques used in breast cancer detection. A data separation/classification method called isotonic separation technique (Ryu et al., 2007) is one of the methods followed in predicting breast cancer. In Salama et al., (2012) different classifiers like multilayer perceptron neural network, combined neural network, probabilistic neural network, recurrent neural network and SVM were analyzed for classification accuracies of breast cancer detection. Lu et al., (2017) proposed an automated computer aided diagnosis framework which consists of ensemble under-sampling (EUS) for imbalanced data processing, the relief algorithm for feature selection, the subspace method for providing data diversity, and Adaboost for improving the performance of base classifiers. They extracted morphological, various texture, and Gabor features for magnetic resonance imaging (MRI).
Wang et al., (2018) proposed an SVM-based ensemble learning algorithm to reduce the diagnosis variance and increase diagnosis accuracy. Sivakumar et al., (2018) developed an algorithm for breast cancer diagnosis based on Supervised Learning in Quest (SLIQ) and Decision Tree algorithms. Peng at al., (2016) proposed an automated breast cancer diagnosis algorithm which organically integrates artificial immune with semi-supervised learning. Jafari-Marandi et al., (2018) presented a data and decision analytic method that employs both supervised and unsupervised learning powers of ANNs to optimize breast cancer diagnosis with regard to decision-making goals. Alwidian et al., (2013) developed a new technique based on a weighted method to select more useful association rules and a statistical measure for pruning rules for breast cancer disease.