Preference risk assessment method based on grey multi index kernel support vector machine model

To improve the precision of sustainable development analysis of leisure agriculture in rural area, a method for the sustainable development analysis of leisure agriculture in rural area based on grey level multi-item kernel SVM model is proposed. Firstly, take such factors as the environmental characteristics, product characteristics, operation level and infrastructure construction of the leisure agriculture into full consideration to build the leisure agriculture evaluation index system and model; secondly, for the “failure” of extreme preference risk evaluation of SVM model, make use of the improved grey level model to preprocess the preference data record and take advantage of the multi-item kernel to improve the SVM algorithm to realize the improvement of the prediction performance of sample data auto-regression model; finally, verify the effectiveness of the method proposed through empirical analysis and give advice and suggestions on the strengthening of the sustainable development of leisure agriculture in rural area.