Projects‎ > ‎

ONR Electronics Tutor


This contract is for the funding of additional options from the Integration of Intelligent Tutoring Systems for Electronics grant. Funded options include: Option 1 (A), Open Source Learning Management System; Option 3 (C), Integrate ElectronixTutor with Components of the Generalized Intelligent Framework for Tutoring; Option 5 (E), Enhanced Question Asking and Answering with Semantic Networks; Option 6 (F), Enhanced Authoring with the BrainTrust System; Option 9 (I), Integration with PAL3 (Personalized Assistant for Life-Long Learning); and Option 10 (J), Assessment and Evaluation.

Title: Integration of ET with Components of Generalized Framework for Tutoring
PI: Xiangen Hu
Co-PIs: Art Graesser, Frank Andrasik, Zhiqiang Cai
Sponsor: ONR
Awarded: $487,048
Dates: 6/1/16-6/30/17

This project will integrate software components and standards of the Generalized Intelligent Framework for Tutoring (GIFT) with PAL3 (Personal Assistant for Life Long Learning), ElectronicTutor (ET), ASSISTments, and any additional e-Sailor course or learning system accessed by PAL3.  Many of the GIFT assets will be implemented in software, whereas those that encounter integration difficulties will be identified in order to stimulate R&D efforts in the future.  The Contractor will assist in integrating PAL3 with three learning environments: (1) ET, (2) one conversation-based intelligent tutoring system, such as an AutoTutor Application Programing Interface (AAPI) to improve reading literacy or mathematics, and (3) one or more courses with conventional computer-based-training that is available and used in the Navy.

Title: Integration with PAL3 (Personalized Assistant for Life-Long Learning)
PI: Benjamin Nye (ICT)
Co-PIs: Arthur Graesser, Zhiqiang Cai, Frank Andrasik
Sponsor: ONR
Awarded: $645,380
Dates: 6/1/16-6/30/17

The Contractor will integrate PAL3 with ElectronixTutor (ET) and at least one other learning environment (such as a course or courses in the e-Sailor program). The University of Southern California’s Institute for Creative Technologies (ICT) will perform the majority of this work, in coordination with the University of Memphis’s Institute for Intelligent Systems (IIS). This integration will allow students to transition seamlessly between different systems over time. Option I has three high-level goals: Content Development: Expanding resources and sharing resources between the systems to cover a broader range of sailor training. This will demonstrate the capabilities of PAL3 and ET to support a wide range of learning and mentorship objectives. Software Integration: Data Standards and Integration: Building data standards between the systems. At the end of this work, the software should be able to merge accounts from both systems and to move between systems while maintaining a synchronized record of progress for the multiple systems. This task will interact with the work in Option C, which will be performing complementary work on incorporating the Generalized Intelligent Framework for Tutoring (GIFT) in ET and other DoD courses to work toward such standards. Student Modeling, Pedagogy, & Task Recommendations: Developing adaptive PAL3 mentorship conversations that unify the user experience across PAL3, ET, and other systems. This will enhance PAL3 by developing adaptive conversational facilities that are sensitive to: (1) a life-long learning record, (2) a sailor’s career goals, (3) performance in courses, and (4) recommended future courses and activities.  This task will be synergistic with Option E.

Title: Enhanced Authoring with the BrainTrust System
PI: Andrew Olney
Co-PIs: Vasile Rus, Art Graesser
Sponsor: ONR
Awarded: $211,023
Dates: 6/1/16-6/30/17

The goal of this project is to improve the authoring process for ElectronixTutor and PAL3 using the BrainTrust system (Olney & Cade, 2016; Olney et al., 2016). In BrainTrust, students read and work with a virtual student on a variety of educational tasks related to the reading (e.g. Navy Electricity and ElectronicsTraining Series). These educational tasks are designed to both improve reading comprehension and contribute to the creation of an ITS based on the material read. After a human student reads a passage, they work with the virtual student to summarize, generate concept maps, reflect on the reading, and predict what will happen next. The tasks and interaction are inspired by reciprocal teaching (Palincsar & Brown, 1984), a well-known method of teaching reading comprehension strategies. The virtual student’s performance on these tasks is a mixture of previous student answers and answers dynamically generated using AI and natural language processing techniques. As the human teaches and corrects the virtual student, they in effect improve the answers from previous sessions, author tutorial dialogues, and improve domain models underlying the ITS.