摘要
研究高校图书馆馆藏质量评价问题。高校图书馆馆藏质量评价是一个多因素、多指标、模糊的非线性过程,传统方法无法进行准确评价,导致评价结果误差较大,准确率较低。为了降低高校图书馆馆藏质量评价误差,提高评价准确率,提出一种立高校图书馆馆藏质量评价模型PSO-SVM。首先构建图书馆馆藏质量评价指体体系,然后将评价指标量化作为支持向量机的输入,图书馆馆藏质量期望输出作为支持向量机输出,并用PSO优化SVM参考,最后对PSO-SVM模型进行仿真。实验结果表明,相对传统馆藏质量评价方法,支持向量机提高了馆藏质量评价准确率,有效降低了评价误差,评价结果更能全面的反映一个高校图书馆馆藏实际情况。
Since university library collection evaluation is a small sample,high-dimension index non-linear process,the traditional evaluation method based on small sample and linear model cannot handle high-dimension index,and may lead to low evaluation accuracy.,A library collection quality evaluation model(PSO-SVM) is put forward based on the particle swarm optimization and support vector machine.The support vector machine is optimized by particle swarm algorithm,and applied to the university library collection quality evaluation.Simulation results show that the compared with traditional evaluation model,the evaluation accuracy of PSO-SVM is higher,and the evaluation result is more scientific and reasonable.It provides a new evaluation method for university library collection evaluatio.
出处
《计算机仿真》
CSCD
北大核心
2011年第6期251-254,共4页
Computer Simulation
关键词
粒子群
支持向量机
图书馆馆藏
高校
Particle swarm optimization(PSO)
Support vector machines(SVM)
Library collection
University