Artificial Neural Network (ANN) with Back-Propagation Algorithm Forecasting Model and Spatiotemporal Visualization for Forestland Rehabilitation


  • Jehan D. Bulanadi
  • Gilbert M Tumibay
  • Mary Ann F. Quioc


For several years, United Nations have been concerned with Global Forests Issues. One of its Sustainable Development Goals focuses on Life on Land emphasizes the importance of forests to people which leads to a vision of increasing the forestland area by 2030. The Philippines, in response to this through the National Greening Program under the Department of Environment and Natural Resources is targeting to rehabilitate 7.1 million hectares of identified unproductive, denuded and degraded forestlands and need to plant 1.5 billion seedlings. This study aims to develop a forecasting model for forestland rehabilitation using Artificial Neural Network with Back-Propagation algorithm. The model will be able to identify among which of the factors or predictors contributed greatly or significantly to the changes occurred in the forestland. Results are then presented using Spatiotemporal Visualization, which illustrates the changes happened in the forestland in a yearly basis using historical data. The model may be used to forecast the size of the forestland that will be rehabilitated for the succeeding years based on the identified predictors, which may be used as a guide by the NGP for reforestation strategic planning and resource management