朱文强.面向O2O服务的移动社交网络个性化可信群体识别模型[J].计算机科学,2018,45(6):76-83
面向O2O服务的移动社交网络个性化可信群体识别模型
Personalized Trustworthy Group Identifying Model Based on O2O Service -oriented Mobile Social Network
投稿时间:2017-05-06  修订日期:2017-08-12
DOI:10.11896/j.issn.1002-137X.2018.06.013
中文关键词:  O2O服务,信任网络,信任模型,可信群体识别
英文关键词:O2O services,Trust network,Trust model,Trustworthy group identifying
基金项目:本文受国家自然科学基金项目(71662014,61602219,61462030)资助
作者单位E-mail
朱文强 江西财经大学软件与通信工程学院 南昌330013 stbrook@aliyun.com 
摘要点击次数: 175
全文下载次数: 129
中文摘要:
      移动通信技术的飞速发展和广泛应用,促进了移动社交网络和O2O服务的高度融合。在订购O2O服务时,人们已习惯先通过移动社交网络咨询O2O服务信息;待服务完成后,再通过移动社交网络发表对O2O服务的体验感受。由于移动社交网络具有开放性和匿名性,用户需要有效识别出可信的用户群体,以对这些服务体验和反馈信息的可靠性进行核实。目前,可信群体识别方面的研究主要集中在云计算和在线社交网络领域,且大多采用全局信任的计算方式,未考虑用户的个性化信任因素,无法适用于面向O2O服务的个性化移动社交网络。针对这一问题,对Advogato信任模型进行扩展,考虑用户的互动程度、社交圈子相似性及兴趣相似性,采用信任容量优先最大流搜索方法建立用户的个性化信任网络,并将识别出的可信用户群体排序输出。基于真实数据集的实验结果表明,该模型在可信群体预测的准确度、漏检率及Top排序范围方面比现有方法的效果更优。
英文摘要:
      With the quick development of mobile communication technique at present,the mobile social network is combining with O2O services more tightly than ever.Users are used to consulting their mobile social networks about those O2O services which they want to order,and submitting their feedbacks and ratings about these services to their mobile social networks after enjoyed these services.However,the mobile social networks are openness,and users can submit their feedbacks anonymously,so it is critical for users to identify the trustworthy group to check if these feedbacks are reliable.The present researches on trustworthy user group identifying are mainly focused on cloud computing and online social networks.Most of these researches use global methods to compute the trust value of users,and neglect factors such as social circles,service preferences,personal interests,and so on,so these existing researches are unfit for the O2O service-oriented personal social networks.To identify the trustworthy group in personal mobile social networks effectively,this paper proposed a trust model based on the famous Advogato model.It takes the interaction frequency,similarity of users’ social circles,similarity of users’ interests into consideration,uses the capacity-first maximum flow search method to transfer the trust flow between users to build their personal trust networks,and finally outputs the ranked trustworthy user group.The experimental results on real dataset show that the proposed trust model has superiorperformance of the prediction accuracy(Pre),missing rate(MsR) and top ranking range(Trr) while comparing to the existing group trust models.
查看全文  查看/发表评论  下载PDF阅读器