社交网络中基于用户投票的推荐机制

发布时间：2018-02-26 02:21

本文关键词： 社交网络推荐机制热点信息个性化好友　出处：《东南大学学报(自然科学版)》2013年02期 　论文类型：期刊论文

【摘要】：为了改善目前社交网络中热点信息推荐与个性化好友推荐的不足,提出基于用户投票的推荐机制.首先,根据众多用户对某条信息的投票情况评估信息的热度与价值,将用户对信息的浏览、评论、转发等操作以及时间因素与用户主动性投票相结合,提出基于用户投票的热点信息推荐算法.然后,根据某个用户对众多信息的投票情况评估用户的兴趣,从用户对网络信息的投票以及浏览情况中提取出用户的兴趣度特征,进而提出基于用户投票的个性化好友推荐算法.最后,针对2个算法进行仿真实验,评估各因素对推荐算法的影响和推荐的有效性.实验结果表明,基于用户投票的推荐机制可以有效地进行热点信息与个性化好友的推荐.
[Abstract]:In order to improve the deficiency of hot information recommendation and personalized friend recommendation in current social networks, a recommendation mechanism based on user voting is proposed. Firstly, the heat and value of information are evaluated according to the voting situation of many users. By combining user's browsing, commenting, forwarding and time factors with user's active voting, a recommendation algorithm of hot information based on user's voting is proposed. The user's interest is evaluated according to a user's voting for a large number of information, and the user's interest is extracted from the user's voting and browsing of the network information. Finally, the simulation experiments of two algorithms are carried out to evaluate the influence of various factors on the recommendation algorithm and the effectiveness of the recommendation. The experimental results show that, The recommendation mechanism based on user voting can effectively carry out the recommendation of hot information and personalized friends.
【作者单位】：东南大学计算机科学与工程学院;上海贝尔股份有限公司;
【基金】：教育部科技发展中心网络时代的科技论文快速共享专项研究资助项目(20110092110053)
【分类号】：TP393.092

【参考文献】