IRES Track I: US-Mexico Collaboration on Multimodal Detection of Objectionable Content in Online Videos in Spanish and English

IRES 轨道 I：美国-墨西哥合作对西班牙语和英语在线视频中的不良内容进行多模式检测

• 批准号: 2106892
• 负责人: Thamar Solorio
• 金额: $ 29.97万
• 依托单位: University of Houston
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-09-01 至 2024-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2106892&HistoricalAwards=false
• 关键词: IRES; Track; US; Mexico; Collaboration

A major form of entertainment for children comes from movies, video games, and online multi-media content. Content from these mass media communications sources has the potential to negatively affect behavior, especially of young viewers. The IRES team will design, implement, and evaluate new multi-modal technologies to detect the presence of content that has been linked to negative effects on viewers while training graduate and undergraduate student in research and professional skills in a multi-cultural, multilingual international environment. The project leverages a long-standing collaboration between faculty at the University of Houston and the Instituto Nacional de Astrofísica, Óptica y Electrónica (INAOE) of Mexico. IRES students will be trained and mentored on multilingual methods of language processing, image/video analysis, and deep learning for multimodal data. The goal of this IRES project is to contribute to the growth of the U.S. national STEM workforce by training a cohort of future computer science professionals (referred to as “IRES Scholars”) to thrive in a multidisciplinary and multicultural environment and enabling them to acquire the necessary knowledge and skills to pursue artificial intelligence research and development. Through this project, IRES Scholars will be exposed to state-of-the-art research in computer vision, natural language processing, and deep learning. They will design, implement, and evaluate new algorithms and systems that will contribute to the crucial goal of providing safer online spaces for young viewers through clear and accurate descriptions of potentially problematic online content. The project employs evidence-based mentoring that will support program sustainability.The objectives of this IRES project are to 1) advance the state-of-the-art in the research field of automated analysis of multimodal content in multilingual and multidomain settings; 2) prepare students with cross-disciplinary and multicultural research skills to join a diverse and global STEM/knowledge workforce; 3) increase enthusiasm, research skills, and professional skills of graduate and undergraduate students from underrepresented population groups in STEM who otherwise might not have an international opportunity for professional development; and 4) immerse students in a community of native Spanish speakers to support engagement in an authentic cross-cultural and multilingual training experience.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.

儿童的主要娱乐形式来自电影、视频游戏和在线多媒体内容。这些大众媒体传播来源的内容可能会产生负面行为，尤其是对年轻观众而言。IRES 团队将设计、实施和评估。新的多模式技术可检测对观众产生负面影响的内容，同时在多文化、多语言的国际环境中培训研究生和本科生的研究和专业技能。该项目利用了双方的长期合作。休斯敦大学和墨西哥国家天文研究所 (INAOE) 的学生将接受语言处理、图像/视频分析和多模态数据深度学习方法的培训和指导。通过培训一批未来的计算机科学专业人员（称为“IRES 学者”）在多学科和领域中蓬勃发展，美国国家 STEM 劳动力的增长通过这个项目，IRES 学者将接触到计算机视觉、自然语言处理和深度学习领域的最先进研究。他们将设计、实施和评估新的算法和系统，通过对潜在问题的在线内容进行清晰、准确的描述，为年轻观众提供更安全的在线空间，该项目将采用基于证据的指导来支持项目的可持续性。 .该 IRES 项目的目标是1) 推进多语言和多领域环境中多模态内容自动分析研究领域的最新技术；2) 培养具有跨学科和多文化研究技能的学生，以加入多元化的全球 STEM/知识队伍； 3) 提高 STEM 中代表性不足的人群的研究生和本科生的热情、研究技能和专业技能，否则他们可能没有国际专业发展机会；4) 让学生融入以西班牙语为母语的社区，以支持参与真正的跨文化和多语言培训体验。该奖项反映了 NSF 的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。