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 109 毫秒
1.
书目推荐旨在建立图书与读者之间的双向连接。新兴网络读书社区中用户的参与互动过程积累了大量用户行为数据,为书目推荐提供了丰富的数据依据。文章针对网络读书社区中的主要用户阅读行为,综合考虑用户评分、阅读状态、标签、时间信息对用户阅读偏好的体现,提出用户阅读偏好程度和偏好方向相结合的用户偏好模型,形成了个性化书目推荐。采集豆瓣读书数据进行模型检验,结果表明,该模型有效地提高书目推荐的准确性和多样性,对网络读书社区实施个性化书目推荐具有参考价值。  相似文献   

2.
<正>如果说上一期专栏我们是得意了一把,侧重分享了Html5页面搭建的可视化新闻策划在社会化传播的优势的话,这一次我们将聊聊现有的可视化新闻产品值得反思的地方。从网易新媒体实验室的操作案例来说,保证用户体验是判断一个可视化新闻成功与否的重要标准,而实现这个标准的关键,是不能徒增用户阅读成本。我们观测了最近制作的30个可视化新闻选题,从PV数、UV数、用户停留时间、用户回流等量化数据进行比较,同时进行了一组20人的用户深度访谈,以进一步分析用户对现有可视化新闻产品的认知和使用偏好,以及挖掘用户对这类新闻策划使  相似文献   

3.
随着移动社交媒体智能推荐功能的发展,智能推荐信息阅读逐渐成为用户移动阅读的重要组成部分。本研究基于扎根理论的研究方法,邀请21位受访者参与半结构化访谈,对访谈资料进行编码分析,提炼了167个初始概念,形成了28个基本范畴,最后归纳为8个主范畴并揭示了各主范畴之间的影响关系,即:智能推荐信息质量、智能推荐系统质量、阅读流体验、心理授权、用户偏好正向影响智能推荐信息阅读,时间约束、任务紧急度负向影响智能推荐信息阅读,智能推荐信息质量、智能推荐系统质量正向影响阅读流体验,智能推荐信息质量正向影响心理授权。同时,本研究还提出了搜索焦点缺失和任务紧急度两个新的基本范畴。研究结果丰富了智能推荐环境下用户阅读行为的研究,为智能推荐信息阅读的发展提供参考。图2。表6。参考文献46。  相似文献   

4.
针对目前移动内容服务系统缺乏自动构建情境服务规则机制的问题,在情境化用户偏好本体模型的基础上,提出量化频繁标引格结构以建立用户内容偏好与情境之间的语义关联,为规则冲突问题和上下文数据可用性问题提供了解决方案。频繁标引格相对于频繁格进一步减少了产生规则所需的结点数目,更便于不同规则的提取和相关参数的计算。设计了频繁标引格分层构建的算法和推荐规则提取的优先级机制,通过实验验证了算法的有效性,并与相关方法进行了比较分析。  相似文献   

5.
基于序列模式的个性化Web页面推荐模型*   总被引:1,自引:1,他引:0  
基于数据挖掘中的序列模式方法,提出一种个性化Web页面推荐模型。该模型首先利用Web使用数据预处理提取Web交易事务集,然后应用序列模式算法挖掘频繁(连续)序列,最后通过构建频繁(连续)序列树生成用户偏好视图以生成个性化Web页面推荐集。  相似文献   

6.
数字阅读时代中国的网络文学深受海外读者的喜爱。本文以海外网络文学平台Webnovel的用户为研究对象,借助用户画像法分析抓取的平台用户信息,构建海外网络文学用户的标签体系,以此研究海外读者的阅读行为与阅读偏好,并提出相应对策,以帮助平台更好地实现精准阅读推荐服务,亦从国家文化发展的战略高度,促进我国对外文化的交流与传播。  相似文献   

7.
针对已有科技文献查询方法往往会遗漏"特征字符不同、但语义相似"的推荐对象的问题,从特征词语义角度研究用户兴趣偏好、科技文献特征的表示方法;提出考虑语义的科技文献与用户兴趣偏好相似性度量方法。在分析个性化科技文献查询推荐业务流程的基础上,设计个性化科技文献查询推荐原型系统ALRS,介绍ALRS的设计思路、功能模块;分析该系统核心算法CGR算法的关键数据结构;详细讨论CGR算法中"基于概念泛化的相似性度量方法"的设计思路,并给出实现该原型系统的核心代码及执行示例。  相似文献   

8.
CSCD个性化服务研究与设计   总被引:1,自引:0,他引:1  
对CSCD个性化服务进行研究,并设计其体系结构和功能模块,包括用户管理、存储服务、推送服务、定制服务和推荐服务;设计数据库结构及其之间的关系;并根据以上设计,采用基于struts2架构,利用hibernate进行服务器的交互,通过tiles整合页面各个模块,应用JQuery架构在JavaScript的编写中,实现CSCD的个性化服务。  相似文献   

9.
基于用户画像的高校图书馆个性化资源推荐服务设计   总被引:1,自引:0,他引:1  
用户画像作为大数据分析背景下个性化推荐服务的设计工具,为高校图书馆领域个性化阅读资源推荐服务提供解决思路。本研究在分析目前个性化推荐和用户画像研究的基础上,引入用户画像技术,从数据基础层、数据处理层、画像构建层、画像服务层设计探讨用户画像的构建流程,重点在用户画像构建和画像服务层面进行阐述,同时从用户基本属性、阅读状态、学习风格、阅读偏好四个维度构建用户多维画像模型,并提出基于冷启动和用户阅读学习过程画像的个性化推荐服务策略,以期为后疫情教育环境下高校图书馆开展个性化资源推荐服务和满足用户多维度阅读学习需求提供参考。  相似文献   

10.
针对科技文献特征词在语义上的层次特性,提出基于概念泛化的内容过滤推荐算法.采用矢量空间模型作为用户兴趣偏好和科技文献特征的描述模型;在比较科技文献特征与用户兴趣偏好的相似程度时,首先从字符层面比较科技文献特征词与用户兴趣特征词,然后在基于ODP目录结构的用户兴趣偏好概念泛化树上对字符不相同的特征词对进行语义比较,并修正特征词权重,以避免遗漏"字符不同,但语义相似"的关键词对.理论分析和实验结果表明,该算法能够更加全面、准确地推荐科技文献对象.  相似文献   

11.
News mentions to academic papers has provided an important channel for research to yield impact on broad audience in the society, and is particularly useful in scholarly communication, scientific outreach and altmetrics. Unlike academic paper, news articles are often influenced by social factors such as culture, ideology and geography. It is therefore uncertain whether news mentions to academic paper is valid for evaluative purpose as expected by altmetrics studies. In response, this study has conducted large scale statistical analysis to explore country's preference over news mentions, based on the assumption that if news mentions to academic paper is not (or weakly) influenced by social factors, no obvious country's preference would be observed, and the vice versa. The major findings are: (1) From the macro perspective, overall distribution of news mentions is highly imbalanced, with several developed countries taking the dominant position. However, no obvious preference towards domestic papers was observed. (2) From the micro perspective, based on the ZINB test results, news from all countries have more positively mentioned domestic papers and simultaneously shown preference over papers from certain countries. (3) In terms of disciplinary comparison, disciplinary differences were observed in the impact strength and specific preference. These results suggest that in micro level country's preference exist for news mentions to academic papers, but in macro level, country's preference is insignificant. Therefore, it is suggested that news mentions can be used for assessing purpose in the macro level.  相似文献   

12.
移动图书馆场景化信息接受内容适配剖析   总被引:1,自引:1,他引:0  
[目的/意义]为提升用户信息接受体验的愉悦度,挖掘用户在不同场景的信息需求期望、信息搜索习惯和信息接受的偏好,并为用户所处场景适配内容是目前业界需要解决的实际问题。[方法/过程]运用微聚合理论对用户情境进行多维度分析,并通过场景识别搭建移动图书馆场景化信息接受内容适配框架。从"场景-内容-情境"适配的视角出发,构建移动图书馆场景化信息接受内容适配机理模型。[结果/结论]以云舟知识服务空间为例,对其信息接受内容适配程度进行分析,规划移动图书馆场景化信息接受的内容适配路径。  相似文献   

13.
个性化关键技术研究综述   总被引:1,自引:0,他引:1  
苏玉召  赵妍 《图书与情报》2011,(1):59-65,98
个性化研究的关键技术主要包括有三个方面的内容:用户建模、推荐系统和评价。首先,用户建模是收集用户数据,目的是为推荐系统提供用户的兴趣、偏好等建立模型;其次,推荐系统的作用是根据用户模型为用户推荐个性化内容,是个性化的核心内容;最后,评价系统根据用户对个性化应用的满意程度,反馈给推荐系统调整建模策略。个性化应用是一个不断反馈修正的过程。  相似文献   

14.
李健  李雪梅 《图书馆论坛》2007,27(2):151-153
数字图书馆的个性化服务是基于定制动态资源中相关用户兴趣的信息检索或过滤即信息收集个性化。由于用户的兴趣是随时变化的,其工作背景也描述了用户访问数字图书馆时当前的信息需求,因此信息收集个性化的重点是要捕获基于用户兴趣和工作背景的信息需求,以提供个性化检索和过滤。  相似文献   

15.
This study aims at understanding international news differences by studying the headlines of over 360,000 news stories posted on the Twitter pages of 12 Arabic and English news organizations. The most referenced countries as well as figures and political actors are examined in these headlines, and the results show that a number of news values elements provide insight into the nature of the news selection. While Arabic channels are mostly focused on the events taking place in the Middle East (proximity), some English-language channels show clear preference for the countries from which they are located, especially CNN and Sky News, as well as Arabic and English state-owned media outlets like France 24 and RT (agenda and ideology). The findings suggest that news content largely follows a number of news values criteria that can explain the news selection process.  相似文献   

16.
Exemplification (the use of examples) in news stories is a common method of providing information about social phenomena to make stories more interesting to audience members. However, previous research has consistently linked exemplification to highly inaccurate perceptions about the prevalence or severity of a given phenomena. The current study further explored outcomes of exemplification by examining the extent to which the practice influenced perceptions of news credibility. Exemplification of public opinion via layperson quotes was found to differentially affect perceptions of news story credibility. Anecdotal exemplification, however, did not predict credibility ratings. Implications for journalism and democratic participation as well as partisans' preference for congenial coverage in news stories are discussed.  相似文献   

17.
This article presents a secondary analysis of two multi-national cross-sectional surveys conducted in 2015 (11 countries, N?=?10,570) and 2017 (4 countries, N?=?2165) to examine the relationship between populist attitudes and media use. The results indicate that populist citizens are more likely to consume news than non-populist citizens. Specifically, populist citizens exhibit a preference for commercial television (TV) news, as well as a tendency to read tabloid newspapers. While they use fewer quality newspapers, public TV news are not systematically avoided. Regarding the online news environment, populist citizens prefer Facebook over Twitter as a source of political information. This selective pattern will be discussed in light of the debates on news audience polarization and political polarization.  相似文献   

18.
ABSTRACT

The online environment has radically changed the way in which users consume, discover and manipulate news. The growing relevance of social media platforms and digital intermediaries for news sharing and consumption increase the likelihood of citizens to be exposed to online news even when they are not seeking it. This digital transformation fundamentally challenges the way online news use and exposure have been conceptualized and measured, affecting also to citizens’ knowledge about public affairs and politics. This article examines the factors that predict the probability to be an “incidentally exposed news user” online. Specifically, based on a representative US sample from the Pew Research Centre, this study analyses the role of media preference, use and trust. Findings indicate that beyond users’ demographics and loyalty, readers’ news preferences, uses and trust, specially of social media platforms, affect their probability to be incidentally exposed to news online. These results have important empirical and theoretical implications for understanding the connection between readers’ news consumption patterns and online exposure, intentional or incidental.  相似文献   

19.
For Latinos in the United States, decisions reflecting which language they prefer for news media consumption can have substantial effects on the quality of the content they receive. However, little is known about what influences Latinos’ language preferences for news media consumption. This project looks to fill that void to the immediate appeal of academics and media analysts by asking, What influences Latinos’ language preference for news? This project builds on social psychology theories highlighting the role played by social identity and self-categorization for inspiring culturally consistent behaviors. This social identity approach focuses on the embrace of Latino characteristics and the complementary nature of Spanish-language media outputs relative to those characteristics. Using the 2006 Latino National Survey, the role of identity is explored. The results indicate that language preferences for news consumption are indeed related to various social identity measures among Latinos in the United States.  相似文献   

20.
对7个典型的国外文献数据库所提供的个性化服务进行调研,着重阐述Web 2.0 环境下个性化服务的5个发展特点。即支持用户在自己的空间内实现知识组织,提供开放机制实现服务的嵌入和融合,提供不打搅用户的行为感知机制进行内容的主动推荐,与社会网络进行集成满足不同用户群体的分享需求,提供基于移动应用的掌上个性化服务。  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