当前位置>主页 > 期刊在线 > 计算机技术 >

计算机技术2019年6期

基于二次邻近算法和支持向量机算法的质量判别算法研究
徐梦琳¹,韩驰²,刘经纬¹
(1. 首都经济贸易大学 信息学院,北京 100070;2. 北京市第二中学,北京 100010)

摘  要:葡萄酒质量的判别是需要聘请专业的葡萄酒专家来品鉴的,因此必然会消耗大量的人力、物力、财力。通过算法来分析葡萄酒的各个属性指标从而确定其质量是十分必要的。但在研究过程中,发现利用常见的多元统计分析算法判别葡萄酒质量时,无论是白葡萄酒还是红葡萄酒均出现了错误率过高的情况。因此,通过有序多分类逻辑回归算法、二次邻近算法、支持向量机算法对葡萄酒质量判别过程中的瓶颈进行了研究。


关键词:多元统计分析;支持向量机算法;二次邻近算法;判别分析



中图分类号:TP274+.3;TP391.1         文献标识码:A         文章编号:2096-4706(2019)06-0104-02


Research on Quality Discrimination Algorithms Based on Quadratic Neighborhood
Algorithms and Support Vector Machine Algorithms
XU Menglin1,HAN Chi2,LIU Jingwei1
(1.School of Information,Capital University of Economics and Trade,Beijing 100070,China;
2.Beijing No.2 Middle School,Beijing 100010,China)

Abstract:Wine quality discrimination is the need to hire professional wine experts to taste,so it will inevitably consume a lot of human,material and financial resources. It is necessary to analyze the various attribute indexes of wine by algorithm to determine its quality. However,in the course of the study,it is found that when the quality of wine is judged by the common multivariate statistical analysis algorithm,the error rate is too high in both white wine and red wine. Therefore,through the ordered multi-classification logic regression algorithm,two proximity algorithm,support vector machine algorithm to study the bottleneck in the process of wine quality discrimination.

Keywords:multivariate statistical analysis;support vector machine algorithm;quadratic proximity algorithm;discriminant analysis


参考文献:

[1] 杨浩,孟娜,王婧,等. 基于支持向量机的京津冀城市群热环境时空形态模拟 [J]. 地球信息科学学报,2019,21(2):190-200.

[2] 房玉林,王华,张莉,等. 不同酿造工艺对毛葡萄酒香气的影响 [J]. 农业工程学报,2007(9):246-250.


作者简介:徐梦琳(1998-),女,汉族,江西南昌人,本科在读,研究方向:大数据分析。