Drug Recommender System Based on Collaborative Filtering for Multiple Sclerosis Patients

Multiple sclerosis is an autoimmune disease that causes physical disability. There is currently no definitive treatment for the disease. Immunosuppressive drugs are used to reduce recurrence and delay disability. The advances in information technology have expanded the use of artificial intelligence systems, including recommendation systems. One of the applications of medical recommender systems is prognosis, diagnosis, and treatment. This study used the data of relapsing-remitting multiple sclerosis (RRMS) patients collected from the MS clinic of Imam Hossein Hospital in Tehran. The data of new patients who are women and aged below 40 years and above 18 years were used. We intend to use clustering and the K-Means method in this study. Also, using cosine similarity, we offer recommendations for a cluster that resembles a new patient. The collaborative filter approach is implemented as one of the recommendation system methods. In other words, a pharmaceutical recommendation system is provided for patients with MS. The results of this study show that the average precision is 98.198%, and the average Recall is 97.756%. Therefore, it performs well for the recommended system.