一种基于数据挖掘的面条品质综合评价方法

为了准确评价面条品质,文章以46种不同品牌的市售面条为材料,通过实验采集面条样本的力学性能数据和其他属性数据,对构建的数据集进行主成分分析,并借助其结果进行聚类分析,样本最终被划分成5个类,对面条样本进行感官评价。参考感官评价的指标再对每一个簇中的样本进行评价,描述出每个簇中面条样本的品质特性。该方法融合了多种数据分析和评价方法,既具备数据分析的客观性,又不失主观评价的优势。

An Integrated Evaluation Approach of Noodles Quality Based on Data Mining

To evaluate the noodles quality accurately,an integrated evaluation approach of noodles quality based on data mining was proposed in this paper.The dataset was composed of noodles mechanical performance data and other attribute data through experiment by using 46 different noodles in various brands as material.Principle Component Analysis（PCA） was used on this dataset.Than clustering analysis was used on this dataset based on PCA＇s result.At last,dataset was divided into 5 clusters.Sense evaluation was used on noodles samples at the same time.Than the quality character of each cluster was described by evaluating noodles samples in each cluster referred to sense evaluate indexes.This approach integrated several data analysis methods and evaluation approaches,which not only had objectivity in data analysis,but also had advantage in sense evaluation.