Sharad Sundararajan

Designing engaging intelligent tutoring systems in an age of cognitive computing

Advances in human computer interaction (HCI) are enabling increasingly more human-like interactions with computers. In this position paper, we explore the impact of several such advances in HCI on the design of an intelligent tutoring system (ITS), with the hypothesis that such systems may drive deeper engagement and hence improve learning outcomes. Researchers have made claims regarding learning gains resulting from self-explanations, frequent addressing of errors and impasses, rich natural language understanding and dialog, appropriate degree of interactivity, and use of multiple representations. However, many studies on current ITSs that have embodied one or more of the above features are showing little to no discernible impact on learning. This is possibly partly due to the poor user experience. Our tutoring system is aimed at addressing learning challenges for K-12 students, by integrating a suite of differentiating technologies around interactivity, dialog, automated question generation, and learning analytics. In this paper, we first review learning theories and insights gleaned from prior research on ITSs that have inspired our design. We then describe the functional architecture of our tutoring system, followed by a preliminary report on the status of the prototype currently being built.

Intelligent Virtual Reality Tutoring System Supporting Open Educational Resource Access.

Virtual Reality is gathering increasing popularity for Intelligent Tutoring Systems. We introduce an approach that improves the baseline VR experience for ITS by enabling access to open educational resources and more intelligent navigation with the support of multiple artificial intelligence algorithms. A preliminary user study result not only reveals the potential of the proposed method, but also helps to identify the clues to improve the current design.

Adaptive Visual Dialog for Intelligent Tutoring Systems.

Conversational dialog systems are well known to be an effective tool for learning. Modern approaches to natural language processing and machine learning have enabled various enhancements to conversational systems but they mostly rely on text- or speech-only interactions, which puts limits on how learners can express and explore their knowledge. We introduce a novel method that addresses such limitations by adopting a visualization that is coordinated with a text-based conversational interface. This allows learners to seamlessly perceive and express knowledge through language and visual representations.

Human-level Multiple Choice Question Guessing Without Domain Knowledge: Machine-Learning of Framing Effects

The availability of open educational resources (OER) has enabled educators and researchers to access a variety of learning assessments online. OER communities are particularly useful for gathering multiple choice questions (MCQs), which are easy to grade, but difficult to design well. To account for this, OERs often rely on crowd-sourced data to validate the quality of MCQs. However, because crowds contain many non-experts, and are susceptible to question framing effects, they may produce ratings driven by guessing on the basis of surface-level linguistic features, rather than deep topic knowledge. Consumers of OER multiple choice questions (and authors of original multiple choice questions) would benefit from a tool that automatically provided feedback on assessment quality, and assessed the degree to which OER MCQs are susceptible to framing effects. This paper describes a model that is trained to use domain-naive strategies to guess which multiple choice answer is correct. The extent to which this model can predict the correct answer to an MCQ is an indicator that the MCQ is a poor measure of domain-specific knowledge. We describe an integration of this model with a front-end visualizer and MCQ authoring tool.

Wizard’s Apprentice: Cognitive Suggestion Support for Wizard-of-Oz Question Answering

Recent advances in artificial intelligence and natural language processing greatly enhance the capabilities of intelligent tutoring systems. However, gathering a subject-appropriate corpus of training data remains challenging. In order to address this issue, we present a system based on a hybrid Wizard-of-Oz technique, which enables cognitive systems to work in tandem with a human operator (the “wizard”), to enhance collection of dialog variants.