مطالعه بر روی مدل رگرسیون بردار پشتیبانی برای پیش بینی سفارش
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|25309||2011||5 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Procedia Engineering, Volume 15, 2011, Pages 1471–1475
The prediction for the order of enterprise is very important. Support vector machine is a kind of learning technique based on the structural risk minimization principle, and it is also a class of regression method with good generalization ability. In this paper, support vector machine is used to model of the prediction for the order. A simulation example is taken to demonstrate correctness and effectiveness of the proposed approach. The selection method of the model parameters is presented.
Production according to the order standing for requirements from customers is the linchpin of improving flexibility and competitive strength of an enterprise. The prediction for the order of enterprise could effectively ahead of the schedule, improve agility and competitive strength, and make the enterprise achieve JIT (Just in Time). Prediction technologies in the field of economics management are numerous. There are two types: qualitative and quantitative method. The qualitative method has the experience to judge the law, the Del's Philippines law, the user survey law and so on. The common applications in quantitative method mainly include regression analysis method, time series method, neural network forecasting method and the grey forecasting method and so on. The return analytic method is high to the historical data material request. The time series method needs to have the massive historical observation value, and has the strong subjectivity and empirical in weighting factor's choice. The neural network forecasting method has the strong auto-adapted ability and learning capability, but requests to the data sample's quantity and the quality high, moreover also has strong empirical in the inserting dimension's determination. The commonly used model in gray forecasting method is GM(1,1) model, and it is suitable in the sequence of strong index rule, but can only describe the monotonous process of change. Support vector machines[4,5] (Support vector machines, SVM) based on structure risk minimum(Structural Risk Minimization, SRM) criterion, overcame the artificial neural networks (Artificial Neural Networks, ANN),which has the shortcoming relied on the experiences of designer, has solved congenital problems well such as dimension, local minimum, small sample of ANN and so on, and given dual attention to the merit. Its topology is decided by support vector. In this paper, SVM theory method is introduced in model of prediction for the order of enterprise, and explained the modeling process by example, confirmed validity of the model through forecasting result.
نتیجه گیری انگلیسی
This article introduces SVR model to order’s Prediction of enterprises, the result of simulation indicated that SVR has characteristics of high modeling precision and strong generalization performance,and confirmed that the Support Vector Regression Model for Order’s Prediction is feasible and effective.