Predicting The Fashion Trends Of Home-textile Based On Big Data

With the new information technology development,the overall direction of the textile industry has accelerated the depth of integration with the technology of the Internet.The level of textile industry informatization promote make textile productss data information growing ever faster,and have the properties of complexity and diversity. A tremendous amount of textile products data facing a shortage of storage and lack of processing capacity situation, resulting in massive data can not be transformed into useful information, performanced that the textile fashion trend forecasting lack the quantitative research. In this paper,Combined the rapid growing domestic textile industry and enterprise data and insufficient analysis processing power,explore to construct a large data processing Hadoop analysis platform,to achieve a quantitative study of the home textile trend forcast.The contents of this paper are as follows:(1) analysis and processing of textile products data based on Hadoop platformConstruction of big data platform based on Hadoop to achieve the analysis of home textile data processing. Using MapReduce method to achieve the original data processing, And separate the popular information data from home textile data,provide data support for the prediction of the trend of home textiles.(2) Home Textiles hue classificationHome textile trends include fabric,color,pattern,style and design content,this paper choose textile popular hue data to predictive.in the case of the color classification standards is different,according to the actual needs, optimize color classification method divided into seven categories,identified seven types hue range of values in different color system.Based on the classification of seven analysis the popular hue quantization.(3) Prediction model of home textile trend forecastUsing trend forecasting method gray system theory to forcast textile fashion trends.In view the error of gray system algorithm for predicting fluctuation data,optimized forecasting model,using smooth processing methods to improve the accuracy of Gray Forecast Model. And,select a different time dimension prediction analysis to determine the optimal prediction model.By model improvement,expanding the application of GM(1,1) model,while the model can take full advantage of the latest data,improve the prediction accuracy.At last,apply the gray model to predict home textile hue trend,to achieve the textile fashion color forecasting high precision,the results of this study provide a reference for textile enterprises to grasp fashion trend information.