Optimizing Cervical Cancer Risk Classification with Feature Scaling Techniques with ANN

Globally, women are affected by cervical cancer as the fourth most frequent malignant condition. Early detection is challenging because its signs are often not noticeable in the disease's early stages. Several factors, including smoking, sexually transmitted diseases, and human papillomavirus (HPV), contribute to the elevated cervical cancer risk. Addressing these danger indicators as well as developing a classification prediction model to determine the presence of the cervical cancer is a significant scientific challenge. This study aims to create a classification model using the Artificial Neural Network (ANN) algorithm, combined with feature scaling. Feature scaling is employed to standardize the data, addressing potential issues arising from differing magnitudes of features. The dataset focuses on two target variables (Biopsy and Schiller) and thirty-two risk factors. The results demonstrate that classification performance is enhanced when the ANN algorithm is applied alongside feature scaling techniques.