关于无线网络用户需求信息快速识别仿真
|
摘要
对无线网络用户需求信息进行快速识别,能够满足网络用户的基本需要,对用户需求信息进行识别,需要利用空间向量模型描述用户需求信息数据,将训练样本输入到支持向量机分类器中进行训练。传统方法将用户需求信息转换为查询词矢量数据,计算相应信息所属领域模型的概率,但忽略了用户信息的不平衡性,导致收敛速度较慢。提出一种基于支持向量机优化的无线网络用户需求信息快速识别方法,采用特征词出现的频率来表示用户信息数据的矢量特征,利用空间向量模型描述用户数据,计算用户数据集中的全部特征词权值完成预处理。将预处理后用户需求信息数据样本进行训练,并采用协同进化的粒子群算法对支持向量机分类器参数进行优化,实现无线网络用户需求信息的快速识别。仿真证明了所提方法能够有效识别出无线网络用户需求信息,且具有收敛速度快的优点。
The fast recognition of user requirement information in wireless network can meet the basic requirements of network users and identify the requirement information. The traditional method ignores the imbalance of user information,leading to the slow convergence. In this paper,we focus on a method to fast recognize user requirement information in wireless network based on support vector machine. The occurrence frequency of feature word was used to denote the vector feature of user information data,and then the space vector model was used to describe the user data and calculate the weight value of all the feature words in user data set. Moreover,the preprocessed data samples of user requirement information were trained. Finally,the particle swarm optimization based on co-evolutionary was used to optimize the parameters of support vector machine classifier. Thus,we could realize the rapid recognition of user requirement information in wireless network. Simulation results show that the proposed method can effectively recognize the requirement information in wireless network,which has fast convergence.
The fast recognition of user requirement information in wireless network can meet the basic requirements of network users and identify the requirement information. The traditional method ignores the imbalance of user information,leading to the slow convergence. In this paper,we focus on a method to fast recognize user requirement information in wireless network based on support vector machine. The occurrence frequency of feature word was used to denote the vector feature of user information data,and then the space vector model was used to describe the user data and calculate the weight value of all the feature words in user data set. Moreover,the preprocessed data samples of user requirement information were trained. Finally,the particle swarm optimization based on co-evolutionary was used to optimize the parameters of support vector machine classifier. Thus,we could realize the rapid recognition of user requirement information in wireless network. Simulation results show that the proposed method can effectively recognize the requirement information in wireless network,which has fast convergence.
引文
[1] 陈新房,刘庆杰,王金峰.基于云计算的地震预警网络系统的设计与实现[J].地震工程学报,2018.40(3):191-198.
[2] 李子华.基于SVD的周期干净信号快速分解[J].电子设计工程,2018,26(6):120-123.
[3] 徐建忠,罗准辰,张亮.语义扩展技术在敏感数据识别中的应用研究[J].现代电子技术,2016,39(12):80-82.
[4] 周勇.浅议客户关系管理中客户需求预测与识别的研究与实现[J].中国管理信息化,2018,12(2):44-45.
[5] 马梧桐,张世颖.数字图书馆用户信息需求影响因素研究[J].情报科学,2016,34(12):92-96.
[6] 刘金.基于数据特征的敏感数据识别方法[J].信息通信,2016,22(2):240-241.
[7] 时明,等.融合生物网络拓扑特性的肿瘤放疗敏感基因信号识别方法[J].模式识别与人工智能,2016,29(12):1057-1064.
[8] 赵小强,等.基于改进FSVM的数据挖掘分类算法[J].兰州理工大学学报,2016,42(2):101-106.
[9] 马晓亭.大数据环境下图书馆敏感数据的识别与保护[J].图书馆论坛,2017,37(4):129-136.
[10] 马莉莉,刘江平.大数据信息传输中恶意攻击数据识别仿真[J].计算机仿真,2017,34(10):375-378.
[2] 李子华.基于SVD的周期干净信号快速分解[J].电子设计工程,2018,26(6):120-123.
[3] 徐建忠,罗准辰,张亮.语义扩展技术在敏感数据识别中的应用研究[J].现代电子技术,2016,39(12):80-82.
[4] 周勇.浅议客户关系管理中客户需求预测与识别的研究与实现[J].中国管理信息化,2018,12(2):44-45.
[5] 马梧桐,张世颖.数字图书馆用户信息需求影响因素研究[J].情报科学,2016,34(12):92-96.
[6] 刘金.基于数据特征的敏感数据识别方法[J].信息通信,2016,22(2):240-241.
[7] 时明,等.融合生物网络拓扑特性的肿瘤放疗敏感基因信号识别方法[J].模式识别与人工智能,2016,29(12):1057-1064.
[8] 赵小强,等.基于改进FSVM的数据挖掘分类算法[J].兰州理工大学学报,2016,42(2):101-106.
[9] 马晓亭.大数据环境下图书馆敏感数据的识别与保护[J].图书馆论坛,2017,37(4):129-136.
[10] 马莉莉,刘江平.大数据信息传输中恶意攻击数据识别仿真[J].计算机仿真,2017,34(10):375-378.