Comparative Analysis of Machine Learning Algorithms and Neural Networks in Early Detection of Breast Cancer
Lokesh Mane, Veera Reddy Aduru, Vamsi Siliveru, Hemanth Kollu, Hiranya Vardhan Gogulamudi · 2024
The fact that breast cancer disproportionately affects women is one of the reasons why it leads to death among a growing number of them. The process of identifying breast cancer is rather time-consuming which makes it crucial to have an appropriate system that can automatically detect the disease when still at its first stage due to lack of adequate resources. Today, Computer-Aided Design (CAD) has been implemented to make the detection of breast cancer easier. However common computer-aided design systems rely on features designed manually, which means they reduce the overall result. Deep learning-based techniques have recently been studied, using machine learning and AI approaches; A lot of machine learning and deep learning algorithms have particularly proven very useful in recent times to discriminate between benign tumors and malignant ones. In contrast to the ML approaches, deep learning and neural networks are better and need less human involvement in pattern recognition schemes. The data used for this analysis is the Wisconsin Breast Cancer Dataset with 569 samples and 32 features. This research aims at a comparative analysis of the results of various machine learning algorithms like Random Forest (RF), Decision Tree (DT), Naïve Bayes and the Feed Forward Neural Network (FNN) and to draw some advantageous insights from using a neural network in the diagnosis of the deadly cancer.