The broader impact of this Small Business Technology Transfer (STTR) Phase I project is to use artificial intelligence methods to help all PhD program stakeholders (students, alumni, faculty, administrators) maximize desired student placements by leveraging available courses and other resources on campus. PhD student placement is a great concern for universities. However, in the absence of data-driven tools that can help administrators track PhD student progress and market needs, there is little that university leaders or faculty can do to continually improve PhD programs and align these programs with the needs of the economy. The total addressable market for AI-driven academic guidance for higher education is estimated at over $1 billion annually. By improving the match between PhD academic preparation and the needs of organizations tackling contemporary challenges in knowledge and technology intensive industries, this project will help universities contribute to society?s grand challenges in areas such as energy, food, disease and transportation. The success of this project will demonstrate the feasibility of continuously gathering adequate data from students, alumni and job postings and using this data to make reliable predictions and actionable individualized recommendations to PhD students that support their academic preparation towards improved market readiness. Education is one of the most important applications of AI, and this project focuses on using AI to empower students, faculty and administrators to maximize the outcomes from the large investments by universities in PhD programs.This Small Business Technology Transfer (STTR) Phase I project aims to collect highly granular data from PhD students, alumni and job market postings and use this data to build prediction and recommendation models to maximize the match between each student?s interests and market needs across long time horizons beyond graduation. While the market expectations for PhD graduate competencies are evolving rapidly and include high levels of multi-disciplinary excellence, PhD programs are evolving slowly, largely due to the lack of data-driven recommendations for appropriate interventions. The proposed R&D plan will develop semi-automated methods for data curation in higher education, then use this data to build novel algorithms using neural network architectures and techniques to predict career outcomes of PhD graduates. The company will also use this data and upstream models to build individualized recommendations using model-based reinforcement learning. The system will suggest the most suitable actions for students, faculty and administrators to maximize the impacts of PhD programs in all disciplines.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.
