Applied Machine Learning Algorithms to Determine Therapy Types for Brain Cancer Patients

Machine learning, a branch of artificial intelligence, is revolutionizing the medical sector by enabling the creation of smart tools and expert systems for data processing. This technology, which uses data to generate predictions or judgments without direct programming, is enhancing productivity and precision in drug discovery and development, revolutionizing scientific research and society. The purpose of this paper is to investigate how to use supervised learning, more specifically classification machine learning algorithms, to discover the best way to classify brain tumor therapy using K-Nearest Neighbor. The result of testing accuracy of 92-94% and training accuracy of 94-96% is that the K-Nearest Neighbor classification technique is incredibly accurate and dependable. It is excellent at predicting age, tumor size, cancer grade, and drug reactions. Precision, recall, and F1 scores, all above 0.95, demonstrate its reliability. However, since accuracy stabilizes with the size of the training set, more data may result in better performance.