Rainfall-induced landslide prediction using backpropagation neural network / Noor Munirah Md Saad

• Main Author: Md Saad, Noor Munirah
• Format: Student Project
• Language: English
• Published: Faculty of Information Technology and Quantitative Sciences 2006
• Subjects: Back propagation (Artificial intelligence); Prediction analysis; Neural networks (Computer science)
• Online Access: http://ir.uitm.edu.my/id/eprint/1736/1/PPb_NOOR%20MUNIRAH%20MD%20SAAD%20CS%2006_5%201.pdf; http://ir.uitm.edu.my/id/eprint/1736/

This research project is about landslide prediction using back propagation neural network. The objectives of the research project are to identify rainfall variable that is used for landslide prediction, apply the back propagation neural network model to classify the risks of landslide and determine whether Back propagation Neural Network can be used as one of prediction tools. A simple three-layer neural network with six input nodes and three output nodes is employed to learn the data. Experiments are performed to determine the optimal learning. The total accuracy of prediction rate is 88.7 %. This research reveals that with a few improvements, back propagation neural network is able to be used in the prediction.