III: Small: Integrated Digital Event Archiving and Library (IDEAL)

III：小型：集成数字事件归档和图书馆 (IDEAL)

• 批准号: 1319578
• 负责人: Edward Fox
• 金额: $ 50万
• 依托单位: Virginia Polytechnic Institute and State University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2013
• 资助国家: 美国
• 起止时间: 2013-09-01 至 2017-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1319578&HistoricalAwards=false
• 关键词: III; Small; Integrated; Digital; Event

The Integrated Digital Event Archive and Library (IDEAL) system addresses the need for combining the best of digital library and archive technologies in support of stakeholders who are remembering and/or studying important events. It extends the work at Virginia Tech on the Crisis, Tragedy, and Recovery network (see http://www.ctrnet.net) to handle government and community events, in addition to a range of significant natural or manmade disasters. It addresses needs of those interested in emergency preparedness/response, digital government, and the social sciences. It proves the effectiveness of the 5S (Societies, Scenarios, Spaces, Structures, Streams) approach to intelligent information systems by crawling and archiving events of broad interest. It leverages and extends the capabilities of the Internet Archive to develop spontaneous event collections that can be permanently archived as well as searched and accessed, and of the LucidWorks Big Data software that supports scalable indexing, analyzing, and accessing of very large collections. Through a new model-based approach to intelligent focused crawling, it improves the quality (e.g., accuracy, coverage, and elimination of noise) of collections of webpages so as to ensure comprehensiveness, balance, and low bias, as is needed for scholarly study of historically important events by social scientists. It incorporates a range of visualization capabilities in support of key stakeholder communities, including archivists, librarians, researchers, scholars, and the general public. IDEAL connects the processing of tweets and webpages, combining informal and formal media, to automatically detect important events, as well as to support building collections on chosen general or specific topics. It supports integration of multiple types and at multiple levels, including key models about the event it is crawling (event models), the sources of information about the event (source models), the mechanisms used for disseminating information about the event (publishing venue models), and the entities related to the event (society /organization models). Integrated services include topic identification, categorization (building upon special ontologies being devised), sentiment analysis, and visualization of data, information, and context.The IDEAL website (http://www.eventsarchive.org) supports searching, browsing, analyzing, and visualizing of event collections (of both tweets and webpages), as well as access to project software, methods, findings, publications, and other results. Usage is encouraged of the integrated system along with a growing number of collections, as well as of particular tools such as for focused crawling, which should aid curators to avoid non-relevant content while including a broader range of sources, improving significantly upon current crawling and archiving methods. Important data and information on events of interest are saved rather than lost, helping preserve our history and culture, in support of public interest, education, policy making, historical analyses, and comparative studies. Students studying sociology, human-computer interaction, digital libraries, information retrieval, computational linguistics, multimedia, and hypertext are gaining experience and contributing in scholarly studies, algorithms, software, interfaces, and big data handling.

集成数字事件档案和图书馆 (IDEAL) 系统满足了将最佳数字图书馆和档案技术相结合的需求，以支持正在记忆和/或研究重要事件的利益相关者。它扩展了弗吉尼亚理工大学危机、悲剧和恢复网络（参见 http://www.ctrnet.net）的工作，以处理政府和社区事件以及一系列重大自然或人为灾难。它满足了对应急准备/响应、数字政府和社会科学感兴趣的人们的需求。它通过抓取和归档广泛感兴趣的事件，证明了 5S（社会、场景、空间、结构、流）方法对智能信息系统的有效性。它利用并扩展了互联网档案馆的功能来开发可永久存档以及搜索和访问的自发事件集合，以及支持可扩展索引、分析和访问超大型集合的 LucidWorks 大数据软件的功能。通过一种新的基于模型的智能聚焦爬行方法，提高网页采集的质量（如准确性、覆盖率、消除噪音等），以保证学术研究所需的全面性、平衡性和低偏差社会科学家对历史上重要事件的研究。它包含一系列可视化功能，为主要利益相关者社区提供支持，包括档案管理员、图书馆员、研究人员、学者和公众。 IDEAL 连接推文和网页的处理，结合非正式和正式媒体，自动检测重要事件，并支持构建所选一般或特定主题的集合。支持多种类型、多层次的集成，包括所爬取的事件的关键模型（事件模型）、事件信息的来源（源模型）、事件信息传播的机制（发布场地模型） ），以及与事件相关的实体（社会/组织模型）。集成服务包括主题识别、分类（基于正在设计的特殊本体）、情感分析以及数据、信息和上下文的可视化。IDEAL 网站 (http://www.eventsarchive.org) 支持搜索、浏览、分析、事件集合（推文和网页）的可视化，以及对项目软件、方法、发现、出版物和其他结果的访问。鼓励使用集成系统以及越来越多的馆藏以及特定工具，例如集中爬行，这将有助于策展人避免不相关的内容，同时包含更广泛的来源，从而显着改进当前的爬行和归档方法。有关感兴趣事件的重要数据和信息得到保存而不是丢失，有助于保护我们的历史和文化，支持公共利益、教育、政策制定、历史分析和比较研究。学习社会学、人机交互、数字图书馆、信息检索、计算语言学、多媒体和超文本的学生正在学术研究、算法、软件、界面和大数据处理方面获得经验并做出贡献。

项目成果

Twitter-Based Knowledge Graph for Researchers

为研究人员提供基于 Twitter 的知识图谱

• 发表时间: 2020-05
• 期刊: Virginia tech
• 作者: Vincent, Kyle;Meno, Emma
• 通讯作者: Meno, Emma

Teaching Natural Language Processing through Big Data Text Summarization with Problem-Based Learning

通过大数据文本摘要和基于问题的学习来教授自然语言处理

• DOI: 10.2478/dim-2020-0003
• 发表时间: 2020-03-01
• 期刊: Data and Information Management
• 作者: Liuqing Li;Jack Geissinger;William A. Ingram;E. Fox
• 通讯作者: E. Fox

Analysis of Moving Events Using Tweets

使用推文分析移动事件

• 发表时间: 2019-07
• 期刊: Virginia tech
• 作者: Patil; Supritha Basavaraj
• 通讯作者: Supritha Basavaraj