Using Different Models of Machine Learning to Predict Attendance at Medical Appointments

摘要: Outpatient absenteeism is a recurring problem worldwide and in Brazil, it chronic problem. The number of appointments exams scheduled not performed, due to the non-attendance patients, reaches significantly high rates can be seen all regions country different types care medical specialties. This practice generates waste resources, disorganizes offer services, limits guarantee at levels assistance. In addition, causes series dissatisfactions from users health system who really need have yet been able access consultations exams. imbalance misuse offer, an increase queue waiting time, as well financial loss since paid for by professional idle absence patients. It necessary understand profile these missing patients try discover reasons that lead this person absent, order predict future consultation. Thus, work presents study machine learning models help whether or patient will attend appointment. As result, end-to-end process was developed, considering exploratory data analysis, pre-processing, creation models, analysis results, deployment most appropriate model web application. found Decision Tree algorithm represents interesting choice final use observations.