无线网络用户信息资源快速识别仿真研究
|
摘要
对无线网络用户资源的快速识别,能够有效提高网络用户信息的整合效率。对用户信息资源的识别,需要通过遗忘函数进行信息资源加权修正,根据修正结果对相似度进行计算,完成用户信息资源的快速识别。传统方法通过决策树对用户信息资源进行多次过滤,对网络信息的决策树进行更新,但忽略了对信息资源的加权修正。提出无线网络用户信息资源快速识别方法。将项目评分相似度和项目类别相似度结合,通过遗忘函数进行加权修正,根据加权修正后得到的评分值对相似度进行计算,反映了用户基本的偏好,提高了快速识别结果的精准度。通过因子分析法对每个项目进行聚类,降低项目向量维数,完成无线网络用户信息资源的快速识别。仿真结果表明,采用所提方法对无线网络用户信息资源进行快速识别时,得到的识别结果精准度较高、识别质量好。
The fast recognition of user resources in wireless network can effectively improve the integration efficiency of network user information. The traditional method updates the decision tree of network information,but ignores the weighted modification for information resources. Therefore,we focus on a method for fast identifying user information resources in wireless network. By combining the similarity of item rating with the similarity of item category,the weighted correction was carried out by forgetting function. Moreover,we calculated the similarity based on the score value obtained after the weighted modification,which reflected the fundamental preference of users and improved the accuracy of fast recognition results. Through factor analysis,we clustered each item to reduce the number of dimensions of item vector. Thus,we could complete the fast identification of user information resources in wireless network. Simulation results show that the proposed method has high accuracy and quality of recognition.
The fast recognition of user resources in wireless network can effectively improve the integration efficiency of network user information. The traditional method updates the decision tree of network information,but ignores the weighted modification for information resources. Therefore,we focus on a method for fast identifying user information resources in wireless network. By combining the similarity of item rating with the similarity of item category,the weighted correction was carried out by forgetting function. Moreover,we calculated the similarity based on the score value obtained after the weighted modification,which reflected the fundamental preference of users and improved the accuracy of fast recognition results. Through factor analysis,we clustered each item to reduce the number of dimensions of item vector. Thus,we could complete the fast identification of user information resources in wireless network. Simulation results show that the proposed method has high accuracy and quality of recognition.
引文
[1]方乐宏,郝文宁,陈刚.基于0-1规划模型的本体不协调解决方法[J].计算机仿真,2017,34(6):260-265.
[2]王晓军.推荐系统中分布式混合协同过滤方法[J].北京邮电大学学报,2016,39(2):25-29.
[3]王占,林岩.基于信任与用户兴趣变化的协同过滤方法研究[J].情报学报,2017,36(2):197-205.
[4]宋楚平.一种改进的协同过滤方法在高校图书馆图书推荐中的应用[J].图书情报工作,2016,60(24):86-91.
[5]冯春卫.基于Zig Bee协议的果园环境信息远程监测系统[J].电子设计工程,2017,16(25):112-116.
[6]吴燎原,贺曦冉,王刚.一种融合社会化信息的改进单类协同过滤方法研究[J].合肥工业大学学报:自然科学版,2016,39(12):1705-1711.
[7]薛朋强,努尔布力,吾守尔·斯拉木.基于网络文本信息的敏感信息过滤算法[J].计算机工程与设计,2016,37(9):2447-2452.
[8]李云玮,马蕾.基于SVM的物联网大数据有效信息过滤挖掘[J].控制工程,2016,23(10):1533-1537.
[9]刘海涛,魏汝祥,袁昊劼.基于互信息的混合属性数据特征选择方法[J].海军工程大学学报,2016,28(4):78-84.
[10]窦羚源,王新华,孙克.融合标签特征和时间上下文的协同过滤推荐算法[J].小型微型计算机系统,2016,37(1):48-52.
[2]王晓军.推荐系统中分布式混合协同过滤方法[J].北京邮电大学学报,2016,39(2):25-29.
[3]王占,林岩.基于信任与用户兴趣变化的协同过滤方法研究[J].情报学报,2017,36(2):197-205.
[4]宋楚平.一种改进的协同过滤方法在高校图书馆图书推荐中的应用[J].图书情报工作,2016,60(24):86-91.
[5]冯春卫.基于Zig Bee协议的果园环境信息远程监测系统[J].电子设计工程,2017,16(25):112-116.
[6]吴燎原,贺曦冉,王刚.一种融合社会化信息的改进单类协同过滤方法研究[J].合肥工业大学学报:自然科学版,2016,39(12):1705-1711.
[7]薛朋强,努尔布力,吾守尔·斯拉木.基于网络文本信息的敏感信息过滤算法[J].计算机工程与设计,2016,37(9):2447-2452.
[8]李云玮,马蕾.基于SVM的物联网大数据有效信息过滤挖掘[J].控制工程,2016,23(10):1533-1537.
[9]刘海涛,魏汝祥,袁昊劼.基于互信息的混合属性数据特征选择方法[J].海军工程大学学报,2016,28(4):78-84.
[10]窦羚源,王新华,孙克.融合标签特征和时间上下文的协同过滤推荐算法[J].小型微型计算机系统,2016,37(1):48-52.