Migraine Categorization based on the Integration of EMD and Naive Bayes Classification

Migraine is a disabling headache that affects over one billion people in the world and is singlehandedly responsible for 42 million years of life lost to disability in 2019. Medical experts need to analyze a patient's medical history and symptoms to determine the type of migraine. It is important to produce a conclusive diagnosis because many types of migraine also have overlapping symptoms with other medical diseases. Making an accurate diagnosis may take a long time as additional medical tests in the complexity of the case. Therefore, this study proposes a computational diagnostic method through a machine learning scheme using Naïve Bayes integrated with entropy minimization discretization (EMD). The proposed method can analyze a patient's symptoms and medical history accurately to reduce the doctor's workload and fasten the diagnosis process. The results show that the Naïve Bayes classifier with EMD achieved high accuracy results; therefore, the proposed method can be a useful method for clinical practice.