Coal Sales Analysis Based On Svm Prediction Systems

In recent years, as China's rapid economic and social development and information technology industries continue to improve, the field of coal sales and network electronics has also made considerable progress. The coal sales data is one of the most important types of data in the coal industry. Analysis and forecasting of such data has become an important routine for coal management department.Support Vector Machine (SVM) as a new machine learning method, because of its small sample study, the quick convergence rate, the strong generalization ability, to be able to obtain the global optimal solution etc, has widely been used in regression forecast fields. This paper studies the theory of support vector machine, and verified the feasibility of using support vector machine model in forecasting coal sales, eventually applied support vector machine model to the analysis and forecast system of coal sales in the actual project, and achieved good results. The researches can be summarized as follows:Firstly, introduced the forecast technique commonly used at home and abroad, and has analyzed the existing methods insufficiency, and with the coal sales data characteristic, proposed the support vector machine time series forecasting technique; Secondly, elaborated the theoretical basis of support vector machine briefly, researched the deep support vector machine solution actual problem's mainstream method, namely the sequence minimum optimization algorithm, and gave the full realization process of the algorithm; Then, support vector machine model applied in the coal sales predict, the experiments showed that the forecast model in aspects of forecast precision and stability display is better, and provided the confirmation support for the system realization; Finally, base on the above research, designed and developed .NET platform coal sales analysis forecast system.Practice has proved that the system can not only carry on the rapid statistical analysis to the coal sales tickets correlation data, and displayed graphically in form that improved the working efficiency, but also implemented coal sales forecast using Support Vector Machine model which offered an important support tool for fast decision-making.