0


Linkify:通过使用链接的开放数据扩充文本来增强阅读体验

Linkify: Enhanced Reading Experience by Augmenting Text Using Linked Open Data
课程网址: http://videolectures.net/iswc2014_yamada_linkify/  
主讲教师: Ikuya Yamada
开课单位: Studio Ousia公司
开课时间: 2014-12-19
课程语种: 英语
中文简介:

我们在阅读报纸、杂志和网页等文本时经常会遇到不熟悉的实体名称(例如,人名或地理位置)。当它发生时,我们通常会执行一系列令人厌烦的操作:选择实体名称,将其提交给搜索引擎,然后通常从网站获取详细信息。在本文中,我们提出了 Linkify,这是一种新颖的工具,它通过自动将实体名称转换为链接并在用户选择链接时显示从链接的开放数据中检索到的实体的概要来增强文本阅读。该工具使用户只需选择链接即可检索实体的信息。此外,为了仅创建对用户有用的链接,我们还开发了一种方法,该方法使用具有广泛特征的机器学习算法来评估实体的有用性。

课程简介: We frequently encounter unfamiliar entity names (e.g., a person’s name or a geographic location) while reading texts such as newspapers, magazines, and web pages. When it occurs, we typically perform a sequence of wearisome actions: select the entity name, submit it to a search engine, and typically obtain detailed information from web sites. In this paper, we propose Linkify, a novel tool that enhances text reading by automatically converting entity names into links and displaying a synopsis of the entity retrieved from Linked Open Data when a user selects the link. The tool enables users to retrieve the information of the entity simply by selecting the link. Further, in order to create only links that are helpful for users, we also developed a method that evaluates the helpfulness of entities using a machine-learning algorithm with a broad set of features. 
关 键 词: 开放数据链接; 机器学习算法; 文本阅读增强
课程来源: 视频讲座网
数据采集: 2021-06-27:zyk
最后编审: 2021-06-27:zyk
阅读次数: 27