Research on early warning analysis model for Karst rocky desertification based on neural network: taking Du'an County of Guangxi as an example

This paper gave a brief of basic principles of neural network, analyzed the BP algorithm applications in the neural network, and introduced the BP algorithm machinery. Based on BP algorithm, takes Karst rocky desertification in Du'an country of Guangxi province as example, proposed the neural network Karst rocky desertification early warning model structure which can realize the rocky desertification warning analysis, calculate the rocky desertification fatalness indexes, and expresses the rocky desertification early warning levels in the thematic map. The result verifies the credibility and the possibility of the early warning analysis model with BP algorithm.