I-Corps: Risk Detection Algorithms for Adolescent Online Safety

这个I-Corps项目的更广泛的影响/商业潜力是，它将通过主动保护青少年免受严重的在线风险而大大造福社会。一个由研究人员、临床医生和行业合作伙伴组成的多学科团队将构建、评估和探索最先进的机器学习算法的商业潜力，这些算法可以在线检测青少年的在线风险行为，包括心理健康问题、性引诱和在线骚扰，以减轻这些风险并预防伤害。该项目将探索将算法发布为1）一个开源项目，创建一个致力于促进青少年在线安全的社区，以及2）一个软件即服务应用程序编程接口，使该解决方案公开提供给社交媒体和基于互联网的公司。其目的是将责任从依赖第三方家长控制软件来保护青少年转移到直接使迎合青少年的在线平台在保护青少年方面发挥更大作用。通过满足这些客户的直接需求，该项目还将通过主动保护青少年免受在线风险，为青少年及其家庭提供次要利益。该I-Corps项目将进一步开发一套针对青少年的算法，不仅可以检测社交媒体数据中的不良内容，还可以识别有问题的行为模式或变化，这些行为模式或变化表明随着时间的推移即将发生的风险。这项工作的智力价值在于多模态数据，机器学习技术和以人为本的方法的新颖和交织的合成，与临床和工业合作伙伴关系相一致，用于设计和开发解决青少年在线风险检测问题的工具。这种方法是变革性的，因为它依赖于通过利用关于社交媒体内容的人类洞察力和元级数据来对青少年社交媒体用户进行更深入的上下文理解，而不是单独分析社交媒体数据。这项工作将使实时在线安全干预，保护和授权青少年互联网用户。这个奖项反映了NSF的法定使命，并已被认为是值得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的支持。

The broader impact/commercial potential of this I-Corps project is that it will greatly benefit society by serving to proactively protect youth from serious online risks. A multidisciplinary team of researchers, clinicians, and industry partners will build, evaluate, and explore the commercial potential for state-of-the-art machine learning algorithms that detect adolescent online risk behaviors online, including mental health issues, sexual solicitations, and online harassment, for the purpose of mitigating these risks and preventing harm. The project will explore the release the algorithms as 1) an open source project that creates a community dedicated to promoting adolescent online safety, and 2) a software as a service application programming interface that makes this solution publicly available to social media and internet-based companies. The intent is to shift responsibility away from relying on third-party parental control software for protecting youth to directly enabling online platforms that cater to youth to play a larger role in protecting teens. By serving the direct needs of these customers, this project will also provide secondary benefits to teens and their families by proactively protecting teens from online risks.This I-Corps project will further develop a suite of algorithms tailored to adolescents that not only detect objectionable content within social media data but can also identify problematic behavioral patterns or changes that are indicative of impending risks over time. The intellectual merit of this work lies in the novel and intertwined synthesis of multi-modal data, machine learning techniques and human-centered approaches, in concert with clinical and industrial partnerships, for the design and development of tools that address the problem of adolescent online risk detection. This approach is transformative because it relies on developing a deeper contextual understanding of teen social media users by leveraging human insight and meta-level data about the social media content, rather than analyzing the social media data by itself. This work will enable real-time online safety interventions that protect and empower teen internet users.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.