Classification Algorithm of Land Quality Based on PLS

Based on modeling idea of partial least squares (PLS) and divided the values of response variable into two classes denoted by 0 and 1, a novel classification algorithm of land quality is set up in this paper. Firstly, the algorithms of multiple linear regression (MLR) and principal component regression (PCR) are introduced and analysed their shortages. Then on the basis of modeling idea of PLS, the classification algorithm of land quality is constructed. The experiment shows that the PLS algorithm doesn´t request distribution of the data, and has best classification pattern ability compared with the algorithms of MLR and PCR. It has more advantages than MLR, PCR, such as simplicity and robustness, clearly qualitative explanation. It is powerful for multicollinearity, particularly when the number of predictor variables is large and the sample size is small, and provides a novel research method for classification of land quality