Date

2016

Department or Program

Computer Science

Primary Wellesley Thesis Advisor

Eni Mustafaraj

Abstract

Educational games and Intelligent Tutoring Systems have been shown to improve student learning outcomes by increasing engagement and providing individualized instruction. However, while introductory programming students frequently benefit from such systems, students in upper-level theoretical courses such as CS 235 (Languages and Automata) have dense textbooks and dry mathematical readings as their primary or only resources. Tutor-Complete aims to fill this gap by presenting two fundamental CS 235 concepts in a game environment. In the first activity, students construct Deterministic Finite-State Automata in order to guide their character across a landscape. In the second activity, students build proofs using the pumping lemma to defeat the “villain” character. Tutor-Complete also fosters a peer-learning environment by encouraging students to explain concepts to each other and providing hints based on past student work. Finally, Tutor-Complete uses Bayesian Knowledge Tracing to model students’ knowledge and tailor the learning experience accordingly.

Share

COinS