Abstract
One-on-one mentoring is effective for helping novices with career development. However, traditional mentoring scales poorly. To address this problem, MentorPal emulates conversations with a panel of virtual mentors based on recordings of real STEM professionals. Students freely ask questions as they might in a career fair, while machine learning algorithms respond with best-match answers. MentorPal is researching rapid development of new virtual mentors, where training data will be sparse. In a usability study, 31 high school students reported (a) increased career knowledge and confidence, (b) positive ease-of-use, and that (c) mentors were helpful (87%) but seldom covered their preferred career (29%). These results demonstrate feasibility for virtual mentoring, but efficacy studies are needed to evaluate its impact, particularly for groups with limited STEM opportunities.
Acknowledgments
The statements and views in this paper are the views of the authors alone. Additionally, many thanks to the USC Armani Lab who organized the EngX STEM fair where MentorPal was exhibited and to the Naval Postgraduate School program which engaged with us on this research. We would also like to extend our thanks to Fred Borgen’s insight into his CAPA inventory items.
Disclosure statement
The authors declare that they have no conflicts of interest in this work.
Additional information
Funding
Notes on contributors
Benjamin D. Nye
Benjamin Nye, Ph.D. is the Director of Learning Science at the University of Southern California, Institute of Creative Technologies. Ben's research tries to remove barriers to development and adoption of adaptive and interactive learning technology so that they can reach larger numbers of learners. Dr. Nye's research has been recognized for excellence in intelligent tutoring systems, cognitive agents, and realistic behavior in training simulations. His research is on scalable learning technologies and design principles that promote learning. He is the membership chair for the International of Artificial Intelligence in Education (IAIED) Society and holds memberships in Educational Data Mining Society (EDM), and Association for the Advancement of Artificial Intelligence (AAAI).
Dan M. Davis
Dan M. Davis is a consultant for USC ICT, focusing on large-scale distributed DoD simulations. At USC's Information Sciences Institute, he was the Director of the JESPP project for JFCOM for a decade. As the Assistant Director of the Center for Advanced Computing Research at Caltech, he managed Synthetic Forces Express, bringing HPC to DoD simulations. Prior experience includes serving as a Director at the Maui High Performance Computing Center and as a Software Engineer at the Jet Propulsion Laboratory and Martin Marietta. He has served as the Chairman of the Coalition of Academic Supercomputing Centers and has taught at the undergraduate and graduate levels. As early as 1971, Dan was writing programs in FORTRAN on one of Seymour Cray's CDC 6500's. He saw duty in Vietnam as a USMC Cryptologist and retired as a Commander, Cryptologic Specialty, U.S.N.R. He received B.A. and J.D. degrees from the University of Colorado in Boulder.
Sanad Z. Rizvi
Sanad Z. Rizvi was a visiting Research Assistant at the Institute for Creative Technologies of the University of Southern California. His interests are in Machine Learning Research and its applications in the real world. Given his Computer Science background, he has accumulated good experience in software development through various research activities and programming projects. His expertise in Python has allowed him to make open-source contributions to that language. Currently, he is focusing on Deep Learning research for Natural Language Processing and is an Instructor for India's prominent MOOC for Natural Language Processing. He earned a Bachelor of Technology degree in Computer Science and Engineering from The National Institute of Engineering, Mysore, India.
Kayla Carr
Kayla Carr is a software analyst for USC ICT. She has worked on PAL3 and MentorPAL, building mobile and web based learning applications for sailors and other learners. She graduated with a Masters in Computer Science from the University of Southern California and started at ICT as a student researcher during her degree.
William Swartout
William Swartout, Ph.D. is Chief Technology Officer and co-founder of the USC Institute for Creative Technologies and a research professor in the Computer Science Department at the USC Viterbi School of Engineering. His research interests include virtual humans, explanation and text generation, knowledge acquisition, knowledge representation, intelligent computer-based education and the development of new AI architectures. In 2009, Swartout received the Robert Engelmore Award from the Association for the Advancement of Artificial Intelligence (AAAI). Swartout is a Fellow of the AAAI, has served on their Board of Councilors, and is past chair of the Special Interest Group on Artificial Intelligence (SIGART) of the Association for Computing Machinery (ACM).
Raj Thacker
Raj Thaker is in the MS in Computer Science program at the University of Southern California. His focus is centered around backend infrastructure development with primary contributions in Java frameworks. He worked with Amazon AWS Internet of Things (IoT) during Summer 2019 working on developing an end-to-end pipeline for data forecasting. During his time at USC, Raj collaborated on research projects with the Institute for Creative Technologies on open-source intelligent tutoring system prototypes. Apart from personal and commercial projects, Raj has worked to provide technology solutions to NGOs like Happy Hearts Foundation and 5050 Leadership and has been a technical judge at coding competitions like Code for Good by JPMorgan Chase.
Kenneth Shaw
Kenneth Shaw was a visiting Research Assistant at the Institute for Creative Technologies of the University of Southern California, selected under the National Science Foundation's Research Experiences for Undergraduates in summer 2018, where he worked on starting a web-enabled version of MentorPal. His current research includes working on safe, efficient human-robot collaboration at Carnegie Mellon University and heterogeneous multi-agent robotics task allocation at Georgia Tech under the Army Research Lab DCIST project. He is currently in his final year studying Computer Engineering and Computer Science in Intelligence at the Georgia Institute of Technology.