PERAMALAN KLB CAMPAK MENGGUNAKAN GABUNGAN METODE JST BACKPROPAGATION DAN CART

• Main Authors: , sulistyowati, , Drs. Edi Winarko, M.Sc., Ph.D.
• Format: Theses and Dissertations NonPeerReviewed
• Published: [Yogyakarta] : Universitas Gadjah Mada 2013
• Subjects: ETD
• Online Access: https://repository.ugm.ac.id/120349/; http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=60373
• Institution: Universitas Gadjah Mada

Forecasting Measles Outbreak in an area is necessary because to prevent widespread occurrence in an area. One way that is done in this study is to predict the incidence of measles by using a combination of backpropagation ANN and CART. Backpropagation ANN is used to predict the incidence of measles periodic data, then the CART method used to perform the determination of an outbreak or non-outbreak area. Backpropagation neural network is one of the most commonly used methods for forecasting which can result in a better level of accuracy than other ANN methods. While the methods of CART is a binary tree method is also popular for the classification, which can produce models or classification rules. Results of this study show that the number of the best window for backpropagation neural network to forecast the outcome affect forecasting accuracy. Methods of the ANN can do forecasting for time series with accuracy 86.71%. The classification using of CART is 88.52%, but the classification with ANN is 83.61%. So that classification was done by CART for prediction outbreak/non outbreak in this research has accuracy more better than classification with ANN backpropagation.