Personalized Curiosity Engine (Pique): A Curiosity Inspiring Cognitive System for Student Directed Learning

Abstract

Pique is an AI-based system for student directed learning that is inspired by a cognitive model of curiosity. Pique encourages self-directed learning by presenting a sequence of learning materials that are simultaneously novel and personalized to learners’ interests. Pique is a web-based application that applies computational models of novelty to encourage curiosity and to inspire learners’ intrinsic motivation to explore. We describe the architecture of the Pique system and its implementation in personalizing learning materials. In exploring the use of Pique by students in undergraduate and graduate courses in Computer Science, we have developed and implemented two computational models of novelty using Natural Language Processing techniques and concepts from recommender systems. In this paper, we describe the Pique model, the computational models for measuring novelty in text-based documents, and the computational models for generating sequences of personalized curiosity-eliciting learning materials. We report the response from students in the use of Pique in four courses over two semesters. The contribution of this paper is a unique approach for personalized learning that encourages curiosity.