Abstract: According to the SVM algorithm theory，SVM based garden design evaluation model is established.By introducing binary PSO algorithm，the characteristic parameters of garden design is chosen to address the dimension disaster caused by a large number of irrelevant or redundant features，the garden design evaluation is maken by SVM multi-classifiers.Case analysis shows that this method can enhance the accuracy and reliability of the garden design evaluation.

Key words: Particle Swarm Optimization（PSO）, Support Vector Machine（SVM）, feature selection, garden design evaluation, multiclassifiers

摘要： 根据支持向量机算法的原理，建立基于支持向量机的园林设计评价模型，通过引入二进制微粒群算法对影响园林设计的特征参数进行选择，解决了大量无关或冗余特征所造成的“维数灾难”和降低分类器性能的问题，利用SVM多类分类器实现了对园林设计的评价。实例分析表明，该方法提高了园林设计评价的准确性和可靠性。

关键词: 微粒群算法, 支持向量机, 特征选择, 园林设计评价, 多类分类器