摘要
由于高校图书馆图书借阅流量具有一定的非线性特性,传统的回归分析、灰色模型等方法难以处理这种非线性时间序列问题,影响了预测精度。为了提高预测精确度,提出粒子群优化RBF神经网络的图书借阅流量预测模型。该方法以图书馆图书借阅流量历史数据进行RBF神经网络建模,采用粒子群算法对RBF神经网络参数进行优化,最后建立了图书借阅流量动态响应模型。预测结果表明该模型预测结果合理,精度较高,为图书馆提高工作效率和服务质量提供了参考依据。
Since the university library borrowing traffic has a certain nonlinear characteristic,the traditional regressionanalysis,grey model and other methods are hard to deal with the nonlinear time series problem,which affects the prediction accuracy.In order to improve the prediction accuracy,a books borrowing flow prediction model based on RBF neural network optimized with particle swarm is proposed.The historical data of books borrowing traffic is used to model the RBF neural network.The particle swarm optimization algorithm is adopted to optimize the parameters of RBF neural network.The dynamic responsemodel of books borrowing flow was established.The prediction results show that the model has reasonable prediction results andhigh prediction accuracy,which provides a reference for the improvement of working efficiency and service quality in library.
作者
陈越华
CHEN Yuehua(Library of Guangxi Teachers Education University,Nanning 530001,China)
出处
《现代电子技术》
北大核心
2017年第19期115-118,共4页
Modern Electronics Technique
关键词
图书借阅
流量
神经网络
粒子群优化
books borrowing
flow
neural network
particle swarm optimization
