QPSO-KM算法在葡萄酒品质分级中的应用  

Application of QPSO-KM Algorithm in Wine Quality Classification

作　　者：邱靖[1] 彭莞云 吴瑞武[3] 张海涛[1] 

机构地区：[1]云南农业大学教务处,云南昆明650201 [2]云南农业大学植物保护学院,云南昆明650201 [3]云南农业大学基础与信息工程学院,云南昆明650201

出　　处：《安徽农业科学》2015年第11期285-286,288,共3页Journal of Anhui Agricultural Sciences

摘　　要：由于影响葡萄酒质量的指标较多,利用主成分分析法,找到了影响葡萄酒质量的指标总计17个。并对两组评酒员的品评数据进行了差异性检验,研究表明,第2组评酒员的评分数据更可信。同时,利用QPSO算法优化KM算法,建立了葡萄酒分类模型。通过试验分析,该算法相对其他两种算法更能搜索到全局最优解,并对葡萄酒样品进行了分级,该算法能处理聚类方面的类似问题。Because of more factors linked to wine quality, using principal components analysis method, 17 factors influencing wine quality were obtained. The difference test was conducted on evaluation data of two group tasters, and it showed that the score of the second group was more reliable than the first group. At the same time, using the QPSO algorithm to optimize the KM algorithm, wine classification model was established. Through the analysis of the experiments, it can search more than the other two algorithms for the globally optimal solution, and the wine samples were classified, the algorithm can deal with similar issues of clustering.

关 键 词：量子粒子群算法 KM算法 葡萄酒酒样 分级模型 

分 类 号：S126[农业科学—农业基础科学]