PSO-SVM模型在葡萄酒品质分类中的应用研究

摘要

葡萄酒作为一种越来越流行的健康饮品,其品质好坏的鉴别分类日益受到关注.基于此研究了一种自适应粒子群支持向量机品质分类模型,并给出了详细的设计步骤和部分Matlab设计代码,最后采用UCI数据库中的wine数据集进行了验证,结果表明,将自适应粒子群算法引入到支持向量机模型参数的选取中,可以减少参数选取的盲目性,提高预测分类的准确性.%The growing popularity of wine as a health drink, its quality is good or bad on the identification received attention. Based on this study an adaptive particle swarm quality of support vector machine classification model, and gives a detailed design steps and design some matlab code, the last wine in the database using UCI packets are verified, the results show that the adaptive PSO algorithm is introduced to the support vector machine model parameters of selection, the selected parameters can reduce blindness and improve the forecast accuracy of classification.