Using Artificial Intelligence to Generate Interventions for Enhancing Student Performance in College STEM Courses

本项目旨在通过设计、开发和评估一个新的平台来提高本科STEM课程的学习成绩，从而为国家利益服务。“来自未来你的信息”（MFAFY）应用程序将通过智能手机及时向学生发送信息，与未来的自己或其他学生选择的化身建立指导关系。这些信息的时间和内容将根据学生的分数、态度、行为和社会互动使用人工智能自动生成。那些面临学习困难、不愿接受正式指导的学生，将受益于这种个性化的授课系统。MFAFY系统将通过提高保留率，提高STEM学位学生的数量和质量，为强大的STEM劳动力做出贡献。应用程序收集的数据将有助于理解本科生面临的经历类型。该项目将综合有关社会经济因素、科学认同、课程评估以及日常态度、行为和社会互动的细粒度多维测量的数据。数据将使用ODIN应用程序收集，该应用程序根据参考时间/日期，位置（GPS）和社交互动（蓝牙）的规则提示学生。数据将被聚类以发现学生“故事”（经验轨迹）的类型学。辅导员将在整个学期中为每种故事类型定制信息。MFAFY应用程序将结合预测故事类型的机器学习模型，以及这些消息。MFAFY将通过随机对照疗效试验进行评估。NSF IUSE: EHR计划支持研究和开发项目，以提高所有学生STEM教育的有效性。通过参与学生学习轨道，该计划支持有前途的实践和工具的创建，探索和实施。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。

This project aims to serve the national interest by designing, developing, and evaluating a novel platform to improve academic performance in undergraduate STEM courses. The “Messages from A Future You” (MFAFY) application will send students just-in-time messages via a smartphone to create a coaching relationship with a future self or other student-selected avatar. The timing and content of these messages will be automatically generated using artificial intelligence based on student scores, attitudes, behaviors, and social interactions. Students who face academic difficulties, and are hesitant to take advantage of formal advising, will benefit from this personalized delivery system. The MFAFY system will contribute to a strong STEM workforce by increasing retention rates, and by improving both the number and quality of STEM degree students. The data collected by the application will add to understanding of the types of experiences faced by undergraduate students. This project will synthesize data on socio-economic factors, science identity, in-course assessments, and fine-grained multidimensional measures of daily attitudes, behaviors, and social interactions. Data will be collected using the ODIN app which prompts students based on rules referencing time/date, location (GPS), and social interaction (Bluetooth). Data will be clustered to discover a typology of student “stories” (experiential trajectories). Counselors will customize messages for each story type, over the semester. The MFAFY app will incorporate machine learning models which forecast story type, together with these messages. MFAFY will be evaluated via a randomized controlled effectiveness trial. The NSF IUSE: EHR Program supports research and development projects to improve the effectiveness of STEM education for all students. Through the Engaged Student Learning track, the program supports the creation, exploration, and implementation of promising practices and tools.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.