Using a Logistic Regression and K Nearest Neighbor Model to Accurately Diagnose Breast Cancer

Abstract

Breast cancer is one of the most dangerous and rapidly growing diseases in the world. Diagnosing breast cancer is expensive, difficult, and time-consuming. However, artificial intelligence and machine learning algorithms can help physicians to diagnose people with breast cancer at an early stage which will help people to avoid exhaustive treatments. The objective of our research was to classify if someone has malignant or benign cancer. We used the Wisconsin Breast Cancer dataset which was obtained from the UCI repository to create models using supervised learning. We used K Nearest Neighbors, and Logistic Regression algorithms to obtain a model with high accuracy. Both the models had an accuracy of 97%. In the future, the model can be enhanced to be more accurate and accessible to people. This research can help others to create models to predict various other cancers. In the future, we would also like to improve the model by using other methods like image recognition and reducing the input from the user to make it more accurate and accessible.