ABSTRACT
The recovery and prediction of Northern American population data, which are closely related to the future development of Northern America and even the whole world, have become significant subjects and captured great attention among sociologists as well as scientists. The historical population data are significant sources for historical overview of Northern America and can reflect the evolutionary history of the continent. The historical data are the comprehensive reflection of the population development under the influence of all factors, and are the significant foundation in population research. The chapter introduces the weights-and-structure-determination (WASD) neuronet model, which is derived from the error back-propagation neuronet. The WASD algorithm can determine directly the weights connecting the hidden-layer neurons and the output-layer neuron, and obtain automatically the optimal structure of Chebyshev-activation neuronet, that is, the optimal number of neurons in the hidden layer. Therefore, Chebyshev-activation neuronet equipped with the WASD algorithm has excellent performance on learning and prediction, and can be used for research.
