The Research And Application About Technology Of Data Mining In Production Decision Of Apparel Sizes And Styles

With the development of electronic commerce and the raise of customer experience, and the pursuit of personalized fashion, the quick responsiveness for consumer demand can be said to be the most prominent essence of the garment industry. The purpose of this research project is to find a scientific and easy method to match clothing shape to different human body, and to set up communication platform between consumers and companies, for providing ideas and enhancing the competitiveness of the enterprise market.First of all, based on the anthropometric massive database of enterprise, this paper extracted aged 20 ~ 30 years’ old young men born in Shanxi Province as the research object. Through the data preprocessing and analyzing the general characteristics by Spss statistical software, a total of 339 people sample data is determined in this study.Second, grey forecast model of other body size, with the body height, bust and waist size of selecting 39 samples, is established by taking advantage of the grey mode and combining with data mining technology. The rest of the 300 samples of other parts of human body size is forecast by the established model. The validity of the established model is verified by posterior deviation test, and the analysis shows that the grey forecast model completes primely the estimation of other body size corresponding key size of the men’s shirts.Finally, based on the body height, chest circumference, waist size of this samples and other parts estimated size of men’s, different somatotype of men’s and apparel sizes and styles are matched by the grey incidence decision. The validity of this method is verified by the qualitative method to compare with enterprise apparel sizes and styles and recommended apparel sizes and styles using it. And then, the apparel sizes and styles of this whole samples are matched by this method. The guidances are provided to put into production for the enterprise, according to the overall analysis of recommended apparel sizes and styles of the whole samples.