关键词: 支持向量机, 粒子群优化, 混沌映射, 嵌入式, 预测, 模型

Abstract: In view of the bad forecasting results of standard-Support Vector Machine (SVM) for product sale series with the normal distribution noise, a SVM based on the Gauss loss function, by name-SVM, was proposed. And then, a short-term intelligent forecasting method for product sales and its parameter-selection algorithm were presented. The results of its application in automobile sale forecasting indicated that the short-term forecasting method based on-SVM was effective and feasible.

Key words: support vector machine, particle swarm optimization, chaotic mapping, embedded, forecasting, models