SBIR Phase II: A Controversy Detection Signal for Finance

SBIR 第二阶段：金融领域的争议检测信号

• 批准号: 1951091
• 负责人: Kaustubh Prabhu
• 金额: $ 74.24万
• 依托单位: Automated Controversy Detection, Inc.
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-05-01 至 2022-10-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1951091&HistoricalAwards=false
• 关键词: SBIR; Phase; II; Controversy; Detection

The broader impact of this Small Business Innovation Research (SBIR) Phase II project will result from development of a technology to automatically detect controversy and disinformation, providing a means for financial institutions to reduce risk exposure. Controversies and disinformation have received public attention and political concern recently. Application of the technology in the finance industry is part of a growing trend toward "alternative data" products relying on artificial intelligence and machine learning, saving analyst time and enabling faster reactions to news stories and social media. Demonstrating the innovation in a quantifiable application, such as finance, is expected to lead to more far-reaching societal impact by enabling users to critically and quantitatively evaluate the often-overwhelming stream of online content.This Small Business Innovation Research (SBIR) Phase II project advances the development of novel algorithms that automatically detect controversy in social media, news, and other outlets. The proposed project will apply a real-time controversy detection signal to financial data, using methods such as language models and machine learning. Additionally, this project will: 1) build core capabilities based on existing controversy detection technology; and 2) construct novel algorithms with broad applicability.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

这项小企业创新研究（SBIR）第二阶段项目的更广泛影响将源于一种自动检测争议和虚假信息的技术的开发，为金融机构提供了一种降低风险的手段。最近，争议和虚假信息引起了公众和政治的关注。该技术在金融行业的应用是依赖人工智能和机器学习的“替代数据”产品日益增长的趋势的一部分，这种趋势节省了分析师的时间，并使他们能够更快地对新闻报道和社交媒体做出反应。在金融等可量化应用中展示创新，有望通过使用户能够批判性地和定量地评估往往势如挡的在线内容流，从而产生更深远的社会影响。这个小企业创新研究（SBIR）第二阶段项目推进了自动检测社交媒体、新闻和其他渠道争议的新算法的开发。拟议的项目将使用语言模型和机器学习等方法，将实时争议检测信号应用于金融数据。此外，本项目将：1)基于现有争议检测技术构建核心能力；2)构建具有广泛适用性的新算法。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Fairness via AI: Bias Reduction in Medical Information

通过人工智能实现公平：减少医疗信息的偏见

• 发表时间: 2021
• 期刊: The 4th FAccTRec Workshop on Responsible Recommendation at RecSys 2021
• 作者: Dori-Hacohen, S.;Montenegro, R.;Murai, F.;Hale, S. A.;Sung, K.;Blain, M.;Edwards-Johnson, J.
• 通讯作者: Edwards-Johnson, J.

Restoring Healthy Online Discourse by Detecting and Reducing Controversy, Misinformation, and Toxicity Online

通过检测和减少网上争议、错误信息和毒性来恢复健康的在线话语

• DOI: 10.1145/3404835.3464926
• 发表时间: 2021
• 期刊: and Toxicity Online
• 作者: Dori-Hacohen, Shiri;Sung, Keen;Chou, Jengyu;Lustig-Gonzalez, Julian
• 通讯作者: Lustig-Gonzalez, Julian