The Application Of Artificial Neural Network In The Determination Of The Weight Of Farmland Classification Factors

Gradation, classification and evaluation of farmland are important parts of land and resources survey, are important foundation work to carry out "Land Administration Law of the People's Republic of China". Comprehensively grasping and scientific quantification of land quality and value not only are advantageous to changing the land resource of government management from quantity primarily to the quantity, quality, ecology coordinating management and scientific use, but also play a vital role of promoting the agriculture, countryside development and farmer wealth, and establishing the resource conservation society.Artificial neural network has the non-linear mapping ability, does not need the precise mathematical model, is good at studying the useful knowledge from the input and output data, is easy to realize the parallel computation and so on. Using neural networks in farmland classification can display fully the merit of the neural networks, reduce subjective influence, and make the result more reasonable. At the same time, artificial neural network provides a new solution, a new thought in farmland classification.The author has studied the basic theory and application method of land evaluation and artificial neural network, elaborated the application and model establishment method of artificial neural network in farmland classification work. At the same time, the author also introduced the realization module of the determination of the weight of farmland classification based on artificial nerve network in detail, contrasted the difference degree between the result of the traditional method and artificial neural network method and the reality situation through the example.