Evidence record · UDINUS · SDGs
- Unit
- Rektorat UDINUS
- Period
- 2010
- Goals
- Record
- 41775969-fd60-4279-8723-2107030d7dde
- Published
- 2026-10-30
Universitas Dian Nuswantoro conducted a comparison of the performances of ARIMA, Neural Network, and Linear Regression models for the prediction of Infant Mortality Rate. The activity involved Purwanto, Eswaran, Chikkannan, Logeswaran, and Rajasvaran as participants. Its scope was based on Infant Mortality Rate data collected in Indonesia during the years 1995-2008.
The comparison used performance measures consisting of Mean Absolute Error, Mean Absolute Percentage Error, and Root Mean Square Error. These measures were used to compare the performance of the different models considered for the prediction of Infant Mortality Rate. The activity therefore focused on model performance using the stated data from Indonesia and the specified evaluation measures.
The results showed that the Neural Network model was the best among the different models considered. The reported configuration used 6 input neurons and 10 hidden layer neurons. It used hyperbolic tangent activation functions for the hidden and output layers. This configuration was identified as the best result in the comparison of the ARIMA, Neural Network, and Linear Regression models.
The activity addresses a health-related research subject through the prediction of Infant Mortality Rate. Universitas Dian Nuswantoro presents this comparison as an institutional research activity involving the stated models, Indonesian data, performance measures, and reported Neural Network result. The available evidence is limited to the aim, data scope, comparison measures, model configuration, and finding described in the source.