Improving medical diagnostics with machine learning: a study on data classification algorithms

This paper investigates the effectiveness of the logistic regression (LR) and random forest (RF) algorithms for classifying breast cancer using the Breast Cancer Wisconsin Dataset, consisting of 699 instances and 10 attributes. After pre-processing the data and performing feature extraction to retain relevant information, the dataset is split into training, validation, and test portions to evaluate the LR and RF algorithms. The LR algorithm achieves an accuracy level ranging from 96% to 97% across different split ratios, and its error rate decreases with larger training sets. The RF algorithm achieves an accuracy level ranging from 96% to 98% across different split ratios. The results indicate that both algorithms are effective for classifying the data, and the figures highlight the impact of different split ratios on accuracy and error rate. Proper selection of the split ratio is essential for obtaining reliable results.