文章摘要
基于PCA-IPSO-SVM的葡萄酒分类研究
Research on wine classification based on PCA-IPSO-SVM
投稿时间:2022-07-02  修订日期:2022-07-02
DOI:
中文关键词: 主成分分析  粒子群算法  支持向量机  葡萄酒分类
英文关键词: principal component analysis  particle swarm optimization  support vector machine  wine classification
基金项目:国家自然科学基金项目(11571315,12101557,11901525);
作者单位邮编
胡青 浙江科技学院 310023
曲润 浙江科技学院 310023
胡珍 湖北工业大学 310023
龚世才 浙江科技学院 310023
摘要点击次数: 29
全文下载次数: 0
中文摘要:
      针对葡萄酒化学数据成分冗余及传统算法分类效果不佳的问题。提出了一种基于主成分分析(principal component analysis,PCA)与改进粒子群算法(improved particle swarm optimization,IPSO)优化支持向量机(support vector machine,SVM)的方法,本方法通过主成分分析法的特征提取能力降低模型输入的维度,再利用改进粒子群算法寻找SVM的最佳参数,从而构建葡萄酒分类模型(PCA-IPSO-SVM)。试验结果表明,改进后的模型不仅消除了变量之间的相关性,充分利用了原始数据的信息价值,有效地提升了预测效率。同时也解决了传统SVM模型分类精度不高以及迭代后期仍存在收敛速度慢的问题。该模型与其他模型相比较,提升了葡萄酒品质分类效率,具有较好的泛化能力。该模型具有一定的适用价值和可操作性,可为酿酒行业的葡萄酒品质分类提供方法借鉴。
英文摘要:
      Aiming at the problems of redundant components of wine chemical data and poor classification effect of traditional algorithms. A method of optimizing support vector machine (SVM) based on principal component analysis (PCA) and improved particle swarm optimization (IPSO) is proposed. This method reduces the dimension of model input through the feature extraction ability of principal component analysis, and then uses the improved particle swarm optimization algorithm to find the best parameters of SVM, so as to build a wine classification model (PCA-IPSO-SVM). The experimental results show that the improved model not only eliminates the correlation between variables, makes full use of the information value of the original data, and effectively improves the prediction efficiency. At the same time, it also solves the problems of low classification accuracy of traditional SVM model and slow convergence speed in the later stage of iteration. Compared with other models, this model improves the efficiency of wine quality classification and has better generalization ability. The model has certain applicable value and operability, and can provide a method reference for wine quality classification in the wine industry.
View Fulltext   查看/发表评论  下载PDF阅读器
关闭

手机扫一扫看 分享按钮