数字化图书档案海量数据快速提取仿真研究  

Research on Rapid Extraction of Massive Data in Digital Book Archives

作　　者：张琅 ZHANG Lang(Library Medical Information Research Center,Jiangxi University of Traditional Chinese Medicine,Nanchang Jiangxi 330004,China)

机构地区：[1]江西中医药大学图书馆医药信息研究中心,江西南昌330004

出　　处：《计算机仿真》2019年第3期397-400,452,共5页Computer Simulation

基　　金：江西省高校人文社会科学研究2015年度项目(TQ1506)

摘　　要：对数字化图书档案海量数据进行快速提取能够提高图书档案数据的利用效率。针对当前数字化图书档案海量数据提取方法存在的提取速度慢,且数据提取质量较差的问题,提出一种基于BP神经网络的数字化图书档案海量数据快速提取方法,利用范围型海量数据属性的值域特征,将数字化图书档案数据的分布样本划分为多个子区间,实现数据分类,并通过构建神经元模型,根据隐含层和输出层的数据输出,确定输出的误差项,调整BP神经网络各层权重,构建基于BP神经网络的数字化图书档案海量数据快速提取模型,实现数字化图书档案海量数据快速提取。实验结果表明,所提方法对数字化图书档案海量数据提取的速度较快,且提取的效率较高,用户对提取结果的满意度较高。At present,the rapid extraction of massive data in digital book archives can increase the utilization efficiency of data,but the extraction speed is slow and data extraction quality is low.Therefore,this article focuses on a method for rapidly extracting massive data in digital book archives based on BP neural network.Firstly,the domain features range-type massive data attributes were used to divide the distribution sample of digital book archive data into many subintervals and thus to realize the data classification.Meanwhile,the neuron model was built.According to the data output at the hidden layer and the output layer,the term of output error was determined and the weight of each layer in BP neural network was adjusted.Finally,the model of rapid extraction for massive data in digital book archives based on BP neural network was built.Thus,we achieved the rapid extraction for massive data in digital book archives.Simulation results verify that the proposed method can fast extract massive data in digital book archives.Meanwhile,the extraction efficiency is high and the degree of satisfaction for the extraction results is high.

关 键 词：数字化 图书档案 海量数据 快速提取 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]