For cultivated land area has small sample data,complex nonlinear characteristic,this paper proposed a cultivated land area method based on least squares support vector machines.Firstly,it selected the influence factors of cultivated land area by correlation coefficient method,then optimized the parameters of least square support vector machines by particle swarm optimization algorithm,lastly built the complex nonlinear model between cultivated area and influence factors.It tested the proposed model by Hunan province cultivated land data,the results show that the proposed model improve the pred...