<![CDATA[Evaluate the best performance comparison of C4.5, Naïve Bayes, K-Nearest Neighbor, Logistic Regression, and Support Vector Machines classification methods for detecting breast cancer using a 10 fold Cross Validation test by comparing the values of accuracy, precision, and recall using confusion matrix . The breast cancer dataset used was 699 records with 11 indicator parameters consisting of Code Number, Clump Thickness, Uniformity of Cell Size, Uniformity of Cell Shape, Marginal Adhesion, Single Epithelial Cell Size, Bare Nuclei, Bland Chromatin, Normal Nucleoli, Mitoses, and Classes obtained from http://archive.ics.uci.edu. The data was processed using Rapid Miner Version 9 software. The results of this study found that the percentage of performance of each classification algorithm analyzed, that is C4.5 Algorithm (accuracy 93.70%, precision 94.26%, recall 87.86%), Naïve Bayes Algorithm (accuracy 96.19 %, precision 92.25%, recall 97.50%), K-Nearest Neighbor Algorithm (95.61% accuracy, precision 94.99%, recall of 92.43%), Logistic Regression Algorithm (accuracy 96.77%, precision 95.93%, recall 94.98%), and Support Vector Machines algorithm (accuracy 96.78%, precision 94.83%, recall 96.20%). The best performance results tested using T-Test found that the Logistic Regression and Support Vector Machines algorithm has the same highest accuracy value that is equal to 0.968.]]>
Copyrights © 2018
