校园无线局域网用户兴趣度算法分析  

ANALYSIS OF USER INTEREST ALGORITHM IN CAMPUS WIRELESS LOCAL AREA NETWORK

作　　者：李旭[1] 刘方爱[1] 刘浩然[1] 

机构地区：[1]山东师范大学信息科学与工程学院,济南250014

出　　处：《山东师范大学学报（自然科学版）》2016年第1期25-30,共6页Journal of Shandong Normal University(Natural Science)

基　　金：国家自然科学基金资助项目（9061200;61572301）;山东省自然科学基金资助项目（ZE2013FM008）

摘　　要：针对目前无线网络用户个性化需求的日益增长，现阶段的个性化校园网络用户服务研究在及时性、稳定性等方面无法满足实际应用的需求。本文提出了一种基于影响条件的无线局域网用户兴趣度矩阵相似度度量算法。首先，利用活跃度筛选出对无线局域网用户行为影响较明显的影响条件，如上网时间、地点等；其次，通过计算异常率进行数据处理，即清除上网行为前后有明显异常的数据；最后，利用本文提出的用户兴趣度矩阵模型，在无线局域网用户间进行相似度计算。实验结果表明，本文提出的用户兴趣度矩阵相似度度量算法在一定程度上提高了用户行为相似度聚类的准确性和有效性，并且减少了无效数据，降低相似度计算的复杂度，能够较好的应用于无线局域网用户行为具体研究中。Focused on the issue that the individual campus network user research in the timeliness,stability and other aspects cannot meet the growing demands,an improved user’s Interest Matrix Similarity Algorithm （IMSA）based on dynamic constraints in wireless LANs was proposed in this paper. Firstly,user’s activity was used to filter the constraints which significantly affected the behaviors,such as time and place. Secondly,the abnormal rate was computed to process the exceptional data. At last,the user’s interest matrix model was proposed to compute mutual similarity of users in wireless LAN. The experimental results show that the proposed algorithm can effectively improve the accuracy of the user’s behavior similarity,by the same time it can reduce the quantity of meaningless data and the computational complexity of similarity,which has better performance in the study of wireless network user behavior.

关 键 词：数字校园 网络用户行为 兴趣矩阵 矩阵聚类 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]