Estimating Breast Cancer Class Using Artificial Neural Network and Logistic Regression Methods

Abstract

In this study, artificial neural networks (ANN) and logistic regression (LR) prediction models for breast cancer type has been developed. The proposed models are implemented with real clinical data for breast cancer type prediction. For purpose of constructing the prediction model, there are 699 instances and 10 attributes used in training and testing for the model. The data used in the ANN and LR models are arranged in a format of 9 input parameters and an output parameter which is the class of breast cancer (benign or malignant). The evaluation was made by comparing the data obtained by the two methods. ANN and LR models have accuracy performance 94,78% and 96,18%, respectively. The LR method's accuracy rate is better than the ANN method. This is because the LR method analyzes by taking into account the form of categorical data.

Investigation of the Effect of Ear Measurements on Sex Estimation in Forensic Sciences Using Machine Learning Techniques: Descriptive Research

“…ANN are a form of computing for predicting a set of outputs from a set of input variables. 31 ANN is made up of neurons connected by numerically weighted links. Each neuron calculates weighted sum of input connections and compares it to threshold value.…”