I-Corps Team: ClaimBuster: Automated, Live Fact-Checking

I-Corps 团队：ClaimBuster：自动化、实时事实核查

• 批准号: 1565699
• 负责人: Chengkai Li
• 金额: $ 5万
• 依托单位: University of Texas at Arlington
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2015
• 资助国家: 美国
• 起止时间: 2015-11-15 至 2017-04-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1565699&HistoricalAwards=false
• 关键词: Corps; Team; ClaimBuster; Automated; Live

New forms of technology and media have made it easier than ever to disseminate false information, which can have adverse consequences on businesses, the financial sector, science, and beyond. Fact-checking is thus critically important to increase accountability, improve discourse, and minimize socioeconomic disruptions. A key challenge is that, in light of the new forms of communications, human fact-checkers have difficulty keeping up with rapidly emerging false data and information. This I-Corps team aims to commercialize ClaimBuster, an automated, live fact-checking system that finds claims, matches them with claims that have already been checked by professionals, and, for new claims not yet checked, provides analysis of relevant data to help check their validity. The team?s technology leverages novel research advances in databases, data and text mining, machine learning, etc. By participating in the I-Corps Teams program, the team?s goal is to clearly identify target customers and their needs, which will enable the team to prepare a viable commercialization plan and further develop necessary components for deploying their technology.

新形式的技术和媒体使传播虚假信息比以往任何时候都更容易，这可能对企业，金融部门，科学等产生不利影响。 因此，事实核查对于加强问责制、改善话语和最大限度地减少社会经济干扰至关重要。 一个关键的挑战是，鉴于新的通信形式，人类事实核查人员很难跟上迅速出现的虚假数据和信息。这个I-Corps团队的目标是将ClaimBuster商业化，这是一个自动化的实时事实检查系统，可以发现索赔，将其与已经由专业人员检查过的索赔进行匹配，并为尚未检查的新索赔提供相关数据分析，以帮助检查其有效性。 团队？的技术利用新的研究进展，在数据库，数据和文本挖掘，机器学习等，通过参与I-Corps团队计划，团队？的目标是清楚地确定目标客户及其需求，这将使团队能够准备一个可行的商业化计划，并进一步开发部署其技术所需的组件。

项目成果

Maverick: A System for Discovering Exceptional Facts from Knowledge Graphs

Maverick：从知识图中发现异常事实的系统

• DOI: 10.14778/3229863.3236228
• 发表时间: 2018
• 期刊: Proceedings of the VLDB Endowment
• 影响因子: 2.5
• 作者: Zhang, Gensheng;Li, Chengkai
• 通讯作者: Li, Chengkai

Maverick: Discovering Exceptional Facts from Knowledge Graphs

Maverick：从知识图中发现特殊事实

• DOI: 10.1145/3183713.3183730
• 发表时间: 2018
• 期刊: Proceedings of the 2018 International Conference on Management of Data
• 作者: Zhang, Gensheng;Jimenez, Damian;Li, Chengkai
• 通讯作者: Li, Chengkai