The application of the spatio-temporal data mining algorithm in maize yield prediction

This paper has advanced a new method of maize yield prediction which is based on the spatio-temporal data mining. On the one hand, this method predicts the sequence of the target object itself by adopting statistics principles. On the other hand, the method calculates the influence of the spatial distribution of adjacent-area soil fertility on maize yield by neural network. Finally, the method obtains the integrated forecast outcome by taking linear regression. Taking the method to predict the maize yield of experimental area for six consecutive years is better than taking no account of the spatial influence. The errors are controlled by 5%. This paper provides a new method to solve the spatio-temporal data update of farmland affected on the maize yield.