大数据资源中用户需求信息定向提取仿真

摘要

对大数据资源中用户需求信息的定向提取,能够有效提高用户资源检索的效率.对用户需求信息的定向提取,需要确定用户需求信息特征函数,构建大数据资源中用户需求信息观测值的实际特征,完成用户需求信息的定向提取.传统方法建立知识本体模型,优化用户需求文档语义标注技术,但忽略了对用户信息观测值的特征进行获取,导致提取精度偏低.提出基于条件随机场(CRFs)模型的大数据资源中用户需求信息定向抽取方法.利用信息空间向量和特征空间向量的内积对大数据资源中每一用户需求信息的特征值进行计算,进而对条件随机场(CRFs)模型进行分析,建立用户需求信息抽取的条件随机场,确定用户需求信息特征函数,得到用户需求信息观测值的实际特征,降低提取的复杂度.实验结果表明,所提方法能够准确对用户需求信息进行分类,保证信息提取效果.%Traditional orientation extraction method of user requirement information in big data resources always ignores to acquire feature of observation data of user information,which leads to poor extraction precision.In order to solve the problem,based on conditional random fields (CRFs) model,a new orientation extraction method is presented in this article.Inner product of information spatial vector and feature space vector is used to calculate characteristic value of each user' s requirement information and the CRFs model is analyzed.Conditional random filed of extraction of the user requirement information is built and characteristic function of the information is confirmed.Actual characteristic of the observation data is obtained and complex rate of the extraction is reduced.Simulation results show that the method can classify the user requirement information and ensure effect of information extraction.