Deciphering Migraine Types: A Machine Learning Odyssey for Precision Prediction

Headache is a common medical complaint affecting people of all age groups. It ranges in severity from mild to severe. Headaches are classified into three types: Tension-type headaches, Cluster headaches, and Migraines. This research study focuses on Migraines which is a severe headaches accompanied by many symptoms and can be caused due to genetic or environmental factors. Migraine is of various types but this research study marks out only 6 types of it. The primary aim of this research is to compare various Machine Learning algorithms to determine the best model with optimal accuracy and precision in predicting the type of migraine based on diverse factors. 6 machine learning algorithms are trained and tested on a machine-generated dataset which contains 23 parameters based on which the algorithm predicts the type of migraine a patient suffers from. The models are finally compared based on 13 metrics or evaluation measures. The model with the best performance is declared to be the one that predicts the type of migraine with the least errors. In this research study, KNN is observed to be the best-performing Machine Learning algorithm with an accuracy and precision of 0.85.