An Associative Based Approach to Analyzing an Online Learning Environment

Bahareh Azarnoush; Jennifer M Bekki; Bianca L. Bernstein; George C Runger

Download Paper | Permalink

Conference: 2013 ASEE Annual Conference & Exposition
Location: Atlanta, Georgia
Publication Date: June 23, 2013
Start Date: June 23, 2013
End Date: June 26, 2013
ISSN: 2153-5965
Conference Session: Emerging Computing and Information Technologies
Tagged Division: Computing & Information Technology
Page Count: 12
Page Numbers: 23.154.1 - 23.154.12
DOI: 10.18260/1-2--19168
Permanent URL: https://peer.asee.org/19168
Download Count: 437

Request a correction

Paper Authors

biography

Bahareh Azarnoush Arizona State University

visit author page

BAHAREH AZARNOUSH is a PhD student at the School of Computing, Informatics, and Decision Systems Engineering, Arizona State University, PO Box 878809, Tempe, AZ 85287-8809, bazarnou@asu.edu. Her research interest is statistical learning.

visit author page

author page

Abstract

An Associative Based Approach to Analyzing an Online Learning Environment Recent years have witnessed a tremendous proliferation of the World Wide Web. This haschanged the way we experience many aspects of everyday life including the ways we communi-cate, conduct business, and educate. Since its advent in the mid-1970s, great advances in educationdelivery through the Web have been made. Indeed, online learning environments (OLE) and effortstowards their improvement have become increasingly common. This paper presents two different approaches for analyzing OLE activity through the mining ofrules. These rules shed light into the users’ learning interests and needs and depict the relationshipbetween different components of the OLE and the resulting learning outcome. The approachesare demonstrated for the existing OLE, CareerWISE (http://careerwise.asu.edu), which is a sitedesigned to teach resilience skills to women students in STEM doctoral programs. The ﬁrst approach involves the discovery of the correlation among different pages of the OLEwhich may be useful in recommending pages that are likely to be of interest to particular users.Rules in the form of {Px , Py , ...} −→ Pz where Pi denotes page i of the OLE are of interest. Suchrules may shed light on the users’ learning needs and interests, and this can be used in recommend-ing Pz to a user upon observing {Px , Py , ...} in his/her active session window. Also, there are typically a large set of navigation paths that can be taken in exploring any web-site. Since an OLE is usually designed with a speciﬁc goal of generating some learning outcomewithin the user, there will be some gain in discovering the navigation paths that are associatedwith achieving this goal. Using this intuition our second approach is based on the discovery of thecorrelation among the page visits and the score on a learning instrument. Such relationships arecaptured in rules in the form of visit to a subset o f pages → target = high. In this approach, it isassumed that an instrument that quantiﬁes learning from the OLE is available and is taken afterexploring the OLE (much like a standard test in the usual educational system). Together, the application of these rules provides guidance on navigating the OLE and allowsusers to reach components of the OLE that may have not been reached otherwise. This, in turn,improves the OLE’s efﬁciency at achieving learning objectives among users. 1

Citation
Format

Azarnoush, B., & Bekki, J. M., & Bernstein, B. L., & Runger, G. C. (2013, June), An Associative Based Approach to Analyzing an Online Learning Environment Paper presented at 2013 ASEE Annual Conference & Exposition, Atlanta, Georgia. 10.18260/1-2--19168

TY - CPAPER
AB - An Associative Based Approach to Analyzing an Online Learning Environment Recent years have witnessed a tremendous proliferation of the World Wide Web. This haschanged the way we experience many aspects of everyday life including the ways we communi-cate, conduct business, and educate. Since its advent in the mid-1970s, great advances in educationdelivery through the Web have been made. Indeed, online learning environments (OLE) and effortstowards their improvement have become increasingly common. This paper presents two different approaches for analyzing OLE activity through the mining ofrules. These rules shed light into the users’ learning interests and needs and depict the relationshipbetween different components of the OLE and the resulting learning outcome. The approachesare demonstrated for the existing OLE, CareerWISE (http://careerwise.asu.edu), which is a sitedesigned to teach resilience skills to women students in STEM doctoral programs. The ﬁrst approach involves the discovery of the correlation among different pages of the OLEwhich may be useful in recommending pages that are likely to be of interest to particular users.Rules in the form of {Px , Py , ...} −→ Pz where Pi denotes page i of the OLE are of interest. Suchrules may shed light on the users’ learning needs and interests, and this can be used in recommend-ing Pz to a user upon observing {Px , Py , ...} in his/her active session window. Also, there are typically a large set of navigation paths that can be taken in exploring any web-site. Since an OLE is usually designed with a speciﬁc goal of generating some learning outcomewithin the user, there will be some gain in discovering the navigation paths that are associatedwith achieving this goal. Using this intuition our second approach is based on the discovery of thecorrelation among the page visits and the score on a learning instrument. Such relationships arecaptured in rules in the form of visit to a subset o f pages → target = high. In this approach, it isassumed that an instrument that quantiﬁes learning from the OLE is available and is taken afterexploring the OLE (much like a standard test in the usual educational system). Together, the application of these rules provides guidance on navigating the OLE and allowsusers to reach components of the OLE that may have not been reached otherwise. This, in turn,improves the OLE’s efﬁciency at achieving learning objectives among users. 1
AU - Bahareh Azarnoush
AU - Jennifer M Bekki
AU - Bianca L. Bernstein
AU - George C Runger
CY - Atlanta, Georgia
DA - 2013/06/23
PB - ASEE Conferences
TI - An Associative Based Approach to Analyzing an Online Learning Environment
UR - https://peer.asee.org/19168
DO - 10.18260/1-2--19168
ER -