Application of discrete orthogonal combinatorial prediction in crime sites

We establish two mathematical models by using Stochastic Partial Elasticity Theory and Artificial Neural Network respectively to obtain two sets of predicted values of of the next crime site. We also establish a prediction model to solve the combinatorial weighting coefficient with prediction precision as optimization index. After combining the two sets of values dynamicly by using Discrete Orthogonal Method, we get a combinatorial predicted value.