移动通信中网络个性化服务信息提取仿真

发布时间：2018-01-27 14:49

本文关键词： 移动通信下社交网络服务提取　出处：《计算机仿真》2016年12期 　论文类型：期刊论文

【摘要】：对移动通信中网络个性化服务信息进行准确提取,可提高移动通信信息服务的质量。进行网络个性化服务信息提取时,需要预测用户对服务项目的未知评分,并计算用户的时间行为信息获得网络用户兴趣相似度,完成网络个性化服务信息提取。但是传统方法通过入挖掘潜在的网络用户社会关系,对移动通信中网络个性化服务信息进行提取,不能精确计算网络用户兴趣相似度,存在信息提取不准确、效率低的问题。提出一种关于用户兴趣度的移动通信中网络个性化服务信息提取方法。上述方法首先将移动通信过程中社交网络与用户评分矩阵组成一个矩阵,融合随机梯度下降理论对该矩阵进行分解,预测用户对服务项目的未知评分,获取相似用户对同一服务项目兴趣度的预测值与真实值之间的误差,并将用户访问时间引入社交网络服务提取过程中,采用用户的时间行为信息获得网络用户兴趣相似度,计算社交网络用户近似度和时间衰弱项,依据计算的结果对网络个性化服务信息进行提取。仿真结果表明,所提方法可以有效提高移动通信中网络个性化服务信息提取精度,网络开销比较稳定。
[Abstract]:The personalized service information in mobile communication are extracted accurately, can improve the quality of mobile communication and information services. Extraction of personalized service of network information, the user needs to predict the unknown on the service item score, access to the network users interest similarity time behavior information and calculate the user's complete extraction of personalized service information network. But the traditional method by the mining potential network users of social relations, the personalized service information in mobile communication network are extracted, user interest similarity cannot be calculated accurately, the existence of information extraction is not accurate, the problem of low efficiency. This paper proposed a method of personalized service of network information on mobile communication user interest degree. The first social network and users mobile communication in the process of scoring matrix composed of a matrix of stochastic gradient descent on the fusion theory Matrix decomposition, prediction of the unknown user service score, similar to the same user access prediction service project interest value and the error between the true value, and the user access time into social network service extraction process, access to the network users interest similarity using the time information of user behavior, calculate the approximate degree and time. Social network users, personalized service of Web information extraction according to the calculation results. The simulation results show that the proposed method can effectively improve the extraction accuracy of personalized service of network information in mobile communication network, the overhead is relatively stable.

【作者单位】：长沙理工大学计算机与通信工程学院;
【基金】：湖南省科学技术计划项目(2011FJ3086)
【分类号】：TN929.5
【正文快照】： _ 1 51胃近年来,随着现代科学计算机技术的迅猛发展,互联网规模的不断扩大,以及移动终端的广泛普及,诸如Facebook、Twitter、Tumblr等社交网络已逐渐成为人们生活中必不可少的一部分[1_2]。人们享受着移动社交网络为广大用户提供的信息传播形式服务[3]。然而信息的迅速传播

【参考文献】