Forecasting the Tuberculosis Morbidity Rate in Indonesia Using Temporal Convolutional Neural Network (TCNN)

摘要

In this modern era, people become aware and use insurance as part of their long-term financial planning in terms of protecting against unpredictable risk. Health insurance is one type of insurance that can be used to help people in alleviating the costs of treatment of the disease they suffered. Some insurance companies offer insurance products that finance the treatment of Tuberculosis (TB), and one of the references used to calculate premiums is TB morbidity. A model for predicting TB morbidity is needed so that premium calculations can be carried out properly. This study aims to predict the TB morbidity rate in Indonesia using the Temporal Convolutional Neural Network (TCNN) method. The results of the model validation were measured by using the Mean Absolute Percentage Error (MAPE) value, where the model produced in this study has a score below 10 %. The model that obtained in this study is then used to forecast TB morbidity rate in Indonesia from 2019 to 2021.