About this paper

Appears in:
Page: 3266
Publication year: 2009
ISBN: 978-84-612-7578-6
ISSN: 2340-1079

Conference name: 3rd International Technology, Education and Development Conference
Dates: 9-11 March, 2009
Location: Valencia, Spain

E-LEARNING GAP ANALYSIS USING AI ALGORITHMS

JM. De Blas, JM. Gutiérrez, JJ. Martínez, R. Barchino

Universidad de Alcalá (SPAIN)
E-learning systems were pushed forward on the mid 90s, as a new kind of distance education model. Since then, these systems have been growing in importance, and now they are the main element in some formative actions. Besides, they also have been growing in complexity, offering more functionality and more ways to help the students. One of these ways to help the students is the competency determination, or gap analysis. A goal of e-learning is to increase efficiency by identifying precisely the training that a student needs, and provide that training in the context of day to day activities of the students.
In the present Learning Management Systems, fixed paths are given to the students. This structure is generally linear. However, it would be useful if the finish point could be established beforehand, and the system then could offer a formative action that led to that point. This method is what we have referred to as gap analysis.
Examining this task, a number of issues arise:
- How to determine which courses or learning materials the user needs to take prior to the desired finish one.
- If multiple solutions are possible, how to choose the optimal.
- In order to offer personalized paths, the users’ already gathered knowledge must be determined.
- How to automate the previous points.
Some of these questions have already been solved, and can be used as a base for our proposed system. Thus, some recent e-learning systems, like the EDVI LMS are capable of tracking the student’s progress and determining this way the students’ knowledge.
On the other hand, in order to create a path, additional metadata that defines links is needed. We propose the use of the SCORM Sequencing and Navigation Specification. Specialized metadata can be attached to the learning materials in order to define its prerequisites and learning outcomes.
The SCORM Sequencing and Navigation Specification was introduced in the SCORM version 2004. It relies on the IMS Simple Sequencing specification and enables the precise declaration of the flow logic involved in controlling the learner’s study path. It provides a more efficient automation of processes like educational resource annotation, and intelligent accessibility management. However, the SCORM only proposes a set of metadata that is attached to the learning materials, is up to the developers how to handle this metadata. Our proposed system uses these metadata to calculate a path of learning materials. In order to do so, we use an approach of constraints, where the constraints are allowed pre-requisites of a learning material, given a specified one.
This problem can be formulated then as a classical Constraint Satisfaction Problem (CSP), and resolved using classical AI techniques. The first approach uses a Backtracking algorithm, a powerful programming language control structure that tries combinations. Our first election of the backtracking algorithm was induced because it has been proved to work well with very large search trees.
Putting all the components together, the proposed system has the following components:
- The backtracking algorithm.
- Learning objects providing data for the SCORM sequencing and navigating metadata.
- A LMS capable of register the user's already learned materials.
In our tests, the system has successfully found the appropriate learning paths rendering the proposed solution viable. We are currently working in an enhanced version using heuristics to decrease computational costs.
@InProceedings{DEBLAS2009ELE,
author = {De Blas, JM. and Guti{\'{e}}rrez, JM. and Mart{\'{i}}nez, JJ. and Barchino, R.},
title = {E-LEARNING GAP ANALYSIS USING AI ALGORITHMS},
series = {3rd International Technology, Education and Development Conference},
booktitle = {INTED2009 Proceedings},
isbn = {978-84-612-7578-6},
issn = {2340-1079},
publisher = {IATED},
location = {Valencia, Spain},
month = {9-11 March, 2009},
year = {2009},
pages = {3266}}
TY - CONF
AU - JM. De Blas AU - JM. Gutiérrez AU - JJ. Martínez AU - R. Barchino
TI - E-LEARNING GAP ANALYSIS USING AI ALGORITHMS
SN - 978-84-612-7578-6/2340-1079
PY - 2009
Y1 - 9-11 March, 2009
CI - Valencia, Spain
JO - 3rd International Technology, Education and Development Conference
JA - INTED2009 Proceedings
SP - 3266
EP - 3266
ER -
JM. De Blas, JM. Gutiérrez, JJ. Martínez, R. Barchino (2009) E-LEARNING GAP ANALYSIS USING AI ALGORITHMS, INTED2009 Proceedings, p. 3266.
User:
Pass: