基于支持向量机的农业科技项目分类研究

Research of Using Support Vector Machine to Carry out Agricultural Technology Projects Classification

农业科技项目投入是解决"三农问题"的关键途径,而农业科技项目分类是农业科技项目投入的参考依据。支持向量机(SVM)是借助最优化方法解决分类问题的方法,较好地克服了"维数灾难"和"有限样本的学习分类"等问题。通过选择不同的核函数和对应的参数可以构造不同的分类器,参数的选择决定了其学习和泛化能力。为此,提出了粒子群优化(PSO)算法和K-折交叉验证来搜索最优参数,并将其应用到农业科技项目分类中。实验结果表明,该方法搜索到的参数达到了较高的准确率,对农业科技项目分类分析有较大的帮助。

The key way to solve the problems of agriculture,rural areas and farmers is to increase the investment of agricultural technology project,and agricultural technology project classification is the reference of agricultural technology project investment.Support Vector Machine（SVM） is an approach to solve classification problem by using optimization method.It can construct different classifiers by choosing different kernel functions and corresponding parameter and parameter＇s choice decides SVM＇s abilities of learning and generalization.To this end,Particle Swarm Optimization（PSO） algorithm and K-fold cross-validation are proposed to search the optimal parameters and take them into the practice of agricultural technology project classification.The result of experiment shows that the parameters searched by Particle Swarm Optimization algorithm and K-fold cross-validation are more precise,which have a great help in analysis of agricultural technology project classification.