Machine Learning Approach to Classify Breast Tissues: A Case Study Using Six-classed Breast Tissue Data

The present study investigates the effectiveness of six Machine Learning (ML) algorithms in classifying breast tissue dataset generated using electrical impedance spectroscopy method. This used available at UCI machine learning repository, consisting 106 spectral records with ten variables. data were partitioned into train and test datasets. Sixty percentage was allocated for balance dataset. Six ML tested accuracy, Cohen’s Kappa, sensitivity specificity. results revealed that backpropagation algorithm (BPN) produced highest accuracy Kappa compared to other six-classed Both Support Vector (SVM) K-Nearest Neighbors (KNN) second-highest Kappa. C5.0 decision tree takes third level. fourth fifth levels are Probabilistic Neural Network (PNN) Quantization (LVQ), respectively. all classes by classification BPN more than eighty percentage, which is higher algorithms. specificity predicted ninety comparatively level Therefore, concludes will effectively classify classed data.