The Research On Data Mining Technology Of Market Forecasting For Clothing Enterprises

With the rapid development of economic globalization and information technology, the market competition is becoming increasingly fierce, the priority to make a reasonable, fast to reflect the enterprise to the market to seize the opportunity, develop. With China’s entry into the WTO market, the clothing industry has been rapid development, but the attendant inventory cost has become an important factor restricting the development of clothing enterprises. Clothing due to the random and other reasons there is a seasonal change and climate prediction, resulting in the production of clothing enterprise complex, most enterprises only through mass production to seize the market, but this only makes the enterprise production cost remains high.The essence of data mining technology is the mathematical modeling, implicit data relation between mass of data the establishment of mathematical model, so as to obtain the optimal solution. At present, the general market forecast method considers only one factor, the established model can not satisfy the more restricted factor, directly affects the accuracy of the prediction results. In recent years the development of neural network theory and perfect for this paper provides a new prediction method. Neural network has very good self-organization, self-learning ability, and now has been applied in many fields of automotive engineering, hydraulic engineering, signal processing, intelligent detection, analysis of credit and market prediction etc..Therefore, this paper put a large amount of historical sales data collection to sorting, extract the useful factors from, using BP neural network technology is used for the forecast of the market, the establishment of the corresponding market forecast model. And through the comparison of the prediction results of three neural networks, multi factor model of BP neural network and time series exponential smoothing model, regression model and other commonly used models, the optimal model.Secondly, the paper discusses the factors affecting the clothing market sales. According to the clothing choice factors determine the impact factor criterion, clothing market paper prediction. Finally, this paper through the use of MATLAB to simulate calculation of the model, the corresponding results are obtained, and do the corresponding description of multi factor BP neural network prediction model’s limitations, are discussed for the further study.