Application of QPSO-KM Algorithm in Wine Quality Classification  

KM算法在葡萄酒品质分级中的应用（英文）

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：《Agricultural Science & Technology》2015年第9期2045-2047,共3页农业科学与技术（英文版）

摘　　要：Since there are many factors affecting the quality of wine, total 17 factors were screened out using principle component analysis. The difference test was conducted on the evaluation data of the two groups of testers. The results showed that the evaluation data of the second group were more reliable compared with those of the first group. At the same time, the KM algorithm was optimized using the QPSO algorithm. The wine classification model was established. Compared with the other two algorithms, the QPSO-KM algorithm was more capable of searching the globally optimum solution, and it could be used to classify the wine samples. In addition,the QPSO-KM algorithm could also be used to solve the issues about clustering.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。

关 键 词：QPSO KM algorithm Wine sample Classification model 

分 类 号：TS262.6[轻工技术与工程—发酵工程]