摘要
运用支持向量机(support vector machine,SVM)和浮点遗传算法相结合的方法对我国专利申请量进行预测。数据仿真显示支持向量机预测方法比人工神经网络和逻辑回归方法有更高的预测精度,结果显示运用浮点遗传算法参数选取的支持向量机方法对我国专利申请量进行预测是可行和有效的。
A forecasting system of patent application quantities is studied by means of applying the support vector machine （SVM） and float-point genetic algorithms. It has higher forecasting precision and stronger generalization ability to forecast applied patent quantities than artificial neural network（ANN） and logistic regression. The results show that the proposed method is feasible and effective.
出处
《运筹与管理》
CSCD
2007年第5期137-141,共5页
Operations Research and Management Science
基金
国家自然科学基金资助项目(70433003)
国家社科基金资助项目(07BJY034)
关键词
预测
模型
支持向量机
专利申请量
forecasting
model
support vector machine （SVM）
patent application quantities
