Breast Cancer Predication Using Data Mining Techniques
As an addendum to the reviewed recent studies on the detection of breast cancer and came to the conclusion that proposed a model to aid in the resolution of the problem of evaluating the severity of the disease's risk, and to learn about the best practices, to reduce time and cost with the aim of improving well-being. We compared three automated methods for blood diseases detection using different method for attribute reduction: PCA (Principle Component Analysis), LDA (Linear Discriminate Analysis), ICA (Independent Component Analysis) ,original data set and six algorithms, which are Naive Bayes, K-Nearest Neighbor, Decision Tree, Logistic Regression, ANN(Artificial Neural Network), and SVM ( Support Vector Machine) for classification. On data set which we have get from kaggle. The outcomes show that in the wake of applying the PCA procedure, the exactness is: 0.909, 0. 87 ,0.91, 0.72, 0.904 and 0.90 for Naive Bayes, Decision Tree, Logistic Regression, SVM, ANN and KNN individually and in the wake of applying the ICA strategy, the exactness is : 0.92, 0.89, 0.93, 0.92, 0.92 and 0.90 for Naive Bayes, Decision Tree, Logistic Regression, SVM, ANN and KNN separately , the precision in the wake of applying the LDA procedure is: 0.92, 0. 90, 0. 90 , 0.925, 0.92 and 0.91 for Naive Bayes, Decision Tree, Logistic Regression, SVM, ANN and KNN separately, while the exactness on the first informational index is: 0.91, 0. 87, 0. 91 , 0.72, 0.91 and 0.91 for Naive Bayes, Decision Tree, Logistic Regression, SVM, ANN and KNN separately. All in all the outcomes gave the idea that when the credulous bayes calculation give the most noteworthy precision subsequent to applying the ICA and LDA strategy which is 92%.
Keywords: Breast cancer; Decision tree; Naïve Bayes; attribute reduction; classification, PCA, LDA
Breast Cancer Predication Using Data Mining Techniques
As an addendum to the reviewed recent studies on the detection of breast cancer and came to the conclusion that proposed a model to aid in the resolution of the problem of evaluating the severity of the disease's risk, and to learn about the best practices, to reduce time and cost with the aim of improving well-being. We compared three automated methods for blood diseases detection using different method for attribute reduction: PCA (Principle Component Analysis), LDA (Linear Discriminate Analysis), ICA (Independent Component Analysis) ,original data set and six algorithms, which are Naive Bayes, K-Nearest Neighbor, Decision Tree, Logistic Regression, ANN(Artificial Neural Network), and SVM ( Support Vector Machine) for classification. On data set which we have get from kaggle. The outcomes show that in the wake of applying the PCA procedure, the exactness is: 0.909, 0. 87 ,0.91, 0.72, 0.904 and 0.90 for Naive Bayes, Decision Tree, Logistic Regression, SVM, ANN and KNN individually and in the wake of applying the ICA strategy, the exactness is : 0.92, 0.89, 0.93, 0.92, 0.92 and 0.90 for Naive Bayes, Decision Tree, Logistic Regression, SVM, ANN and KNN separately , the precision in the wake of applying the LDA procedure is: 0.92, 0. 90, 0. 90 , 0.925, 0.92 and 0.91 for Naive Bayes, Decision Tree, Logistic Regression, SVM, ANN and KNN separately, while the exactness on the first informational index is: 0.91, 0. 87, 0. 91 , 0.72, 0.91 and 0.91 for Naive Bayes, Decision Tree, Logistic Regression, SVM, ANN and KNN separately. All in all the outcomes gave the idea that when the credulous bayes calculation give the most noteworthy precision subsequent to applying the ICA and LDA strategy which is 92%.
Anahtar Kelimeler: Breast cancer; Decision tree; Naïve Bayes; attribute reduction; classification, PCA, LDA