基于支持向量机的图书馆借阅量预测

Library loan prediction based on support vector machine

为了准确预测图书馆借阅量,本文提出基于支持向量机的图书馆借阅量时序预测方法,支持向量机能有效解决非线性、高维、小样本等问题,并采用遗传算法选取合适的支持向量机训练参数,以此增加支持向量机的泛化能力。首先提出了支持向量机预测模型,并进行实例分析,将华北科技学院图书馆流通部1997-2007年度借阅量作为本文的实验数据。实验结果表明支持向量机的图书馆借阅量预测效果优于径向基(Radial Basis Function,RBF)神经网络。

In order to predict library loan correctly, library loan prediction method based on support vector machine is presented in the paper. Support vector machine can solve the problems of nonlinear, ,small sample,and genetic algorithm is employed to select the suitable parameters of support vector machine,thus,the generalization ability of support vector machine is improved.Firstly,the prediction method based on support vector machine is introduced,then,the case is analyzed. Annual library loan of library circulation department of north China institute of science and technology is used as the experimental data of the paper. The experimental results demonstrate that library loan prediction results of support vector machine are better than those of RBF neural network.