Abstract

ExplaNet is a web-based, anonymous, asynchronous explanation-sharing network. Instructors post questions to the network, and students submit explanatory answers. Students then view and rank all explanations submitted by their peers before optionally resubmitting a final and revised answer.

Our thesis is that by using ExplaNet to submit answers and view peer-authored explanations, students can improve their comprehension and retention. We believe ExplaNet harnesses the power of face-to-face collaborative learning in an on-line environment, In addition, we believe that a small subset of explanations is the most beneficial to each individual student; we believe that we can select this subset by analyzing student characteristics and preferences.

To test these theses, we performed four in-class evaluations. We found that students who viewed peer-authored explanations between submitting explanations showed greater improvement in submission scores and retention than students who did not. Students stated that the process of viewing explanations was helpful and that they learned more from using ExplaNet than they would from traditional methods of assessment. We implemented a variation of the naive Bayesian classifier to select a subset of explanations for each student. Our algorithm was able to successfully predict which answers students preferred based on their learning preferences and background characteristics.

Citation

Masters, J. ExplaNet: A learning tool and hybrid recommender system for student-authored explanations. Ph.D. thesis, University of California, Santa Cruz. Retrieved August 16, 2024 from https://www.learntechlib.org/p/116761/.

This record was imported from ProQuest on October 22, 2013. [Original Record]

Citation reproduced with permission of ProQuest LLC.

For copies of dissertations and theses: (800) 521-0600/(734) 761-4700 or https://dissexpress.umi.com

Keywords

• computer science
• Technology of Education