首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献,搜索用时 62 毫秒
1.
一种结合推荐对象间关联关系的社会化推荐算法   总被引:2,自引:0,他引:2  
随着社会化媒体的兴起,信息资源的数量呈现爆炸式增长,如何在海量的信息中帮助用户发现有用的知识成为亟需解决的问题.社会化推荐方法作为一种有效的信息过滤技术,由于能够结合社会网络的特点,模拟现实社会中的推荐过程,在分析用户历史行为的基础上,主动向用户推荐满足他们兴趣和需求的信息,受到了研究者们的广泛关注.但目前已有的方法大都只从用户间社会关系的角度出发,仅认为相互信任的朋友间具有相似的兴趣爱好,而忽略了推荐对象间的关联关系对推荐结果产生的影响.针对以上存在的问题,文中从推荐对象间关联关系的角度出发,假设具有关联关系的推荐对象更容易受到同一用户的关注,并进而在已有的社会化推荐算法的基础上,提出了一种结合推荐对象间关联关系进行推荐的算法.算法使用共享的潜在特征空间对目标函数的求解过程进行约束,使其在考虑用户间社会关系的同时,也考虑到推荐对象间关联关系所起到的重要作用.实验结果表明,与主流的推荐算法相比,文中所提出的方法在分类准确率和评分误差等多种评价指标上都取得了更好的结果.  相似文献   

2.
熊丽荣  刘坚  汤颖 《计算机科学》2016,43(9):255-260, 265
利用移动设备上下文、移动社会化网络等信息进一步提高推荐系统的预测准确率,并缓解可能存在的数据稀疏性和冷启动问题,已经成为移动推荐系统的主要任务。采用基于矩阵分解的因子分析方法,结合用户、服务和用户社会化网络信息进行服务推荐,可以缓解数据稀疏性和冷启动问题;同时,为了增加信任矩阵密度,引入间接信任关系,提出了一种符合移动社会化网络特点的信任度计算方法,该方法仅利用移动社会化网络结构信息构建信任矩阵,从而减少用户对信任关系的主动标识。实验结果表明,引入间接信任关系能够提高预测精度,同时 比传统的协同过滤算法和已有的一些矩阵分解方法具有更好的预测准确率,特别是在评分数据稀疏的情况下。  相似文献   

3.
通过构建一个能够体现不同用户-项目交互信息和项目属性信息的图网络,将推荐问题转化为异构图网络的节点嵌入问题并提出一种用于生成用户-项目表示的联合学习算法.在该联合学习算法中,采用带权的个性化排序算法体现用户对于不同项目的喜好程度的差异,融合项目信息拓展用户的行为模式,更好地找到用户的喜好.实验结果表明,该算法在多种不同的推荐评价指标上比基准算法具有明显的优势.  相似文献   

4.
社会网络中包含大量的社会信息,如何从这些社会信息中发掘对用户有用的信息已成为学者和专家的研究热点。本文提出一种基于社会正则化的推荐算法:把改进的矩阵分解技术应用到社会化推荐中;利用社会网络中用户间的朋友关系来优化对用户的建模,学习更好的用户特征空间模型;利用社会网络中的标签信息建立用户和物品的关系,并利用这种关系来优化用户-物品的建模。实验结果表明,改进后的推荐算法的精确度高于传统的推荐算法,有效地解决了社会信息冗余问题。  相似文献   

5.
协同主题回归(CTR)将概率矩阵分解和主题模型结合应用于推荐系统,在许多推荐应用中取得了成功,但该模型没有考虑用户社会关系对用户兴趣的影响。针对该问题,引入概率链接函数来评估社会关系网络对用户兴趣的影响,并以此约束目标函数。在CTR的基础之上,提出一种融入用户社会关系的协同主题回归模型(USRCTR),结合用户项目评分信息、项目内容和社会关系网络,构建一个基于分层贝叶斯模型的推荐引擎。在Lastfm数据集上实验表明,与其他几种CTR改进方法对比,该模型的训练时间更短,推荐精度更高。  相似文献   

6.
基于用户及其历史行为间关系的社会化推荐引起了广泛的关注,而现在的推荐模型几乎都忽略了社会关系的异质性和多样性。针对该不足,提出个人和社会隐含因子(ISLF)联合模型来进行社会化推荐的方法。该方法结合协同过滤和社交网络建模方法来进行社会化推荐,利用最新的混合隶属随机分块模型为每个用户提取相应的社会因子向量;并采用一种优化的期望最大化算法(EM算法)来优化ISLF模型,以便进行最快的期望计算;最后,基于真实的数据集豆瓣网来进行实验。结果表明该方法比现存的社会化推荐方法提供的推荐更准确,质量更高,效果更好。  相似文献   

7.
针对现有的基于LBS(Location Based Service)个性化推荐系统在构建用户兴趣模型时存在的缺陷,提出一种综合LBS和社会网络标签的个性化推荐(LTCF)方法。通过引入网络标签和用户社会关系,从用户标注的标签资源中找到拥有共同兴趣爱好的用户关系以及从社会网络中找到与目标用户关系紧密的用户,同时结合考虑用户兴趣爱好随空间不断变化的特点,依据协同过滤算法,计算用户社会关系度和用户空间相似性,依此得到目标用户的最近邻集合,在最近邻集基础上给出推荐结果。实验结果表明,相比于传统的基于LBS推荐方法,LTCF模型在查全率和产准率有了显著的提升,能更好地反映出用户偏好,显著提高了推荐准确度。  相似文献   

8.
许彪 《软件工程师》2014,(11):42-45
由于手机中GPS系统定位精度低和室内环境中卫星信号弱,造成移动SNS在室内和近距离环境中社交网络扩展性能差。针对这种情况,提出了基于蓝牙技术的近距离社交网络扩展算法。算法首先在服务器端保存SNS用户手机蓝牙地址与用户的对应信息,然后按照同一区域内"一台蓝牙手机搜索到另一台蓝牙手机"的蔓延模式找到直接相邻的周边用户列表,再在服务器端对周边用户列表按照深度优先算法进行社交网络的深度搜索,找到尽可能多的周边用户并建立社交网络。实验结果表明,移动SNS系统采用该算法比采用现有技术能更加有效、快速地扩展近距离范围内的社交网络。  相似文献   

9.
多数基于标签的推荐算法都存在推荐方式单一的问题,没有充分利用社会关系等其他信息。针对这一问题,提出了一种融合信任关系、时间因子和标签信息的个性化推荐算法TTLMF,该算法在现有基于标签的个性化推荐算法的基础上,充分利用了用户之间的信任关系和当前上下文的时间信息,使得推荐项目更加符合用户的需求。在公共数据集last.fm上进行了实验,结果表明TTLMF算法在准确率、召回率、[Fmeasure]以及覆盖率这四个指标上具有更好的推荐效果,一定程度上缓解了数据稀疏性和用户的冷启动问题。  相似文献   

10.
研究表明在社会网络推荐中添加明确的社会信任明显提高了评分的预测精度,但现实生活中很难得到用户之间明确的信任评分。之前已有学者研究并提出了信任度量方法来计算和预测用户之间的相互作用及信任评分。提出了一种基于Hellinger距离的社会信任关系提取方法,通过描述二分网络中一侧节点的f散度来进行用户相似度计算。然后结合用户分组信息,将提取的隐式社会关系加入改进的概率矩阵分解中,提出一种新的基于用户组群和隐性社会关系的概率矩阵分解算法(CH-PMF)。实验结果表明,提出的模型与应用实际用户明确表示的信任分数推荐结果表现几乎相同,且在无法提取到明确信任数据时,CH-PMF有着比其他传统算法更好的推荐效果。  相似文献   

11.
Social fMRI: Investigating and shaping social mechanisms in the real world   总被引:1,自引:0,他引:1  
We introduce the Friends and Family study, a longitudinal living laboratory in a residential community. In this study, we employ a ubiquitous computing approach, Social Functional Mechanism-design and Relationship Imaging, or Social fMRI, that combines extremely rich data collection with the ability to conduct targeted experimental interventions with study populations. We present our mobile-phone-based social and behavioral sensing system, deployed in the wild for over 15 months. Finally, we present three investigations performed during the study, looking into the connection between individuals’ social behavior and their financial status, network effects in decision making, and a novel intervention aimed at increasing physical activity in the subject population. Results demonstrate the value of social factors for choice, motivation, and adherence, and enable quantifying the contribution of different incentive mechanisms.  相似文献   

12.
As the penetration of mobile phones in societies increases, there is a large growth in the use of mobile phones especially among the youth. This trend is followed by the fast growth in use of online social networking services (SNS). Extensive use of technology can lead to addiction. This study finds that the use of SNS mobile applications is a significant predictor of mobile addiction. The result also shows that the use of SNS mobile applications is affected by both SNS network size and SNS intensity of the user. This study has implications for academia as well as governmental and non-for-profit organizations regarding the effect of mobile phones on individual’s and public health.  相似文献   

13.
Social network sites (SNSs) and mobile phones are becoming increasingly important in teenagers’ lives. Using data collected from a nationally representative survey (N = 800), this study explores the variation of social capital by SNS adoption, different SNS activities, and mobile personal talk among teenagers. The results indicate that SNS adoption and mobile personal talk can not only enhance teenagers’ close ties with friends, but also jointly promote teenagers’ civic engagement. Among SNS users, mobile personal talk also increase teens’ network capital. Different SNS activities such as commenting on friend’s Facebook pictures and joining Facebook groups have different relationships with social capital, and such relationships are moderated by mobile personal talk.  相似文献   

14.
Mobile devices such as tablets, smart phones, and portable computers are connecting users in a myriad of contexts. Social networks can be a benefit for individuals in the knowledge acquisition process, and group dynamics disclosed by Social Network Analysis provides a good basis for studying how mobile collaboration is affected by social networks. This study focuses on two aspects of social network mechanism, namely eigenvector centrality and network reciprocity. The assumption is that social networks influence mobile technology use behaviors, and can be optimized to leverage mobile collaboration. A series of controlled field experiments involving 327 groups were conducted to evaluate the differences between two modes of mobile collaborative settings: individual non-collaborative vs. group collaborative. Results indicate that by leveraging social networks in a mobile platform, study participants were able to positively heighten their collaborative knowledge acquisition process through enhanced group interactions and enjoyment.  相似文献   

15.
This paper defines the concept of social weight as a design consideration and presents the e-SUIT, a social weight research platform incorporated covertly within a traditional business suit. The e-SUIT allows its user to strike a balance between a given technology's derived benefit and its social consequence. As the e-SUIT is designed for research within a business context, it is built upon commercially available enterprise software. This work is a first step towards subjecting the empirical social interactive phenomena of wearable technology to quantitative design analysis. Proof of concept testing shows access to commercially available enterprise applications with a distinct, user selectable, strata of social weight.  相似文献   

16.
Clinicians interested in taking a proactive approach to healthy cancer survivorship might consider the use of a social networking and videosharing platform tailored specifically for young adult cancer survivors. This study examines six key factors that may influence a childhood cancer survivor’s participation in a social networking and videosharing intervention program tailored to their needs: (1) the individual’s social capital, defined as resources accessed by individuals through a broad range of social connections, (2) social support, (3) family interaction, (4) self-efficacy, (5) depression, and (6) self-reported quality of life. Fourteen healthy childhood cancer survivors participated in a social networking and videosharing intervention program, LIFECommunity, over a period of 6 months. Young adult cancer survivors with weak “bonding” social capital with other cancer survivors, little social support from friends and family, and lower family interaction participated in the social networking intervention more than those with stronger social capital and larger bases of support. The findings suggest that cancer survivors used the social network as a way to fulfill needs that were not being met in their “offline” lives. The study provides a deeper understanding of the factors that contribute to the success of social networking interventions for young cancer survivors.  相似文献   

17.
移动社会性网络服务引领了网络发展的新时代,是以帮助人们建立社会性网络为目的的互联网应用服务,受到了广大用户的喜爱。但是这些网站繁多,用户登录所有的网站不太方便。针对社会性网络,运用种子扩展的算法对网络用户所有的社交关系进行归纳分析,构造出"好友圈子",为用户提供社区服务。对某大学的产品调研表明好友圈子为用户提供了便捷、聚合性的服务,为商家寻求盈利模式提供了新的关注焦点,具有较高的商业价值。  相似文献   

18.
19.
Nowadays, we are immersed in the social and mobile networks era. As a positive consequence of this, collaborative and mobile learning in educational environments have been encouraged thanks to the use of computing for human learning. By coupling the advantages of collaborative and mobile learning, the teaching-learning processes involved in postgraduate courses may be greatly enhanced. The pedagogical experiences in this regard lived by the authors in the Alpha–Beta Research Group when coupling collaborative and mobile learning in the context of postgraduate level courses, are presented in this paper.  相似文献   

20.
This study has four primary purposes: to investigate the level of interactivity with 15 sources of political information, determine the degree of reliance on each of the sources, assess perceptions of credibility, and compare the strength of interactivity to the strength of reliance on judgments of credibility. This study includes sources that have received little attention in the credibility literature such as social network sites, Twitter, and mobile device applications. Respondents interact with and rely on the sources moderately. Credibility ratings range from moderate to highly credible. Reliance predicts credibility of 14 of 15 sources as compared to interactivity that predicts credibility of just 9 of 15 sources. Interactivity with sources that are inherently collaborative (e.g. Twitter, social media, talk radio) more strongly predicts credibility than interactivity with sources that are more source-to-user based (e.g. CNN, political websites).  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号