Files in this item

FilesDescriptionFormat

application/pdf

application/pdfSUBRAHMANYAM-THESIS-2017.pdf (465kB)
(no description provided)PDF

Description

Title:Retain: building a concept recommendation system that leverages spaced repetition to improve retention in educational settings
Author(s):Subrahmanyam, Shilpa
Advisor(s):Zhai , ChengXiang
Department / Program:Computer Science
Discipline:Computer Science
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:M.S.
Genre:Thesis
Subject(s):spaced repetition
education
concept recommendation system
Abstract:There is a glaring lack of focus on long-term retention in today's educational paradigms. Moreover, research in the area of learning, memory, and specifically, promoting long-term retention has produced several robust and experimentally validated principles. A lot of this work can be leveraged to place some much-needed emphasis on long-term retention in educational settings. One such principle is spaced repetition -- a technique that has been empirically proven to promote long-term retention. The applications of current spaced repetition algorithms are limited to atomic concepts -- concepts that don't have any conceptual dependencies. In order to apply current spaced repetition formulae to more general contexts, we need to develop a system that can take conceptual dependencies into account. In this paper, we propose a framework called Retain that does exactly this. Retain is a system that can be used in virtually any educational context -- not just contexts that solely involve atomic concepts (i.e. learning vocabulary terms). It is a concept recommendation system that provides students with suggestions about when to review various concepts based on their understanding of parent concepts and the principle of spaced repetition. The results produced by Retain as well as the rules upon which Retain was built were evaluated by a group of teachers and were overwhelmingly favored over other concept recommendation baselines.
Issue Date:2017-04-26
Type:Thesis
URI:http://hdl.handle.net/2142/97490
Rights Information:© 2017 by Shilpa Subrahmanyam. All rights reserved.
Date Available in IDEALS:2017-08-10
Date Deposited:2017-05


This item appears in the following Collection(s)

Item Statistics