Abstrait

Developing Methods To Train Neural Networks For Time-Series Prediction In Environmental Systems

Yongliang Shi, Liang Chen, Jin Liu, Yidong Lei

We present a novel method to training neural networks for predicting future variable values of environmental system. Time-series data including soil, streamwater and climatic variables were measured hourly over several month periods in two situations in Qingpu district, 45 kilometers west Shanghai city, using data loggers and other measuring instruments. The data sets were used to train neural networks using three different methods, including a novel, biologically plausible system. Temporal pattern recognition capabilities using each method were investigated. The novel method proved equally capable in predicting future variable values using large data sets as the other two methods. An argument is made for this method, named the ‘Local Interaction’ method, providing valid competition to other neural network and statistical methods in the detection of patterns and prediction of events in complex environmental systems.