=Paper=
{{Paper
|id=None
|storemode=property
|title=New Web Tool to Create Educational and Adaptive Courses in an E-Learning
Platform Based Fusion of Web Resources
|pdfUrl=https://ceur-ws.org/Vol-867/Paper34.pdf
|volume=Vol-867
|dblpUrl=https://dblp.org/rec/conf/icwit/ChaouiL12
}}
==New Web Tool to Create Educational and Adaptive Courses in an E-Learning
Platform Based Fusion of Web Resources==
https://ceur-ws.org/Vol-867/Paper34.pdf
New Web tool to create educational and adaptive courses
in an E-Learning platform based fusion of Web resources
Mohammed Chaoui1, Mohamed Tayeb Laskri2
1,2
Badji Mokhtar University – Annaba, Algeria
1
chaoui.mohamed@yahoo.fr, 2laskri@univ-annaba.org
Abstract. The evolution of new communication and information technologies
led to a very high rate of innovation in online education. This opens doors for
several major research projects at universities, institutes and research centers,
all over the world. The content of training courses and quality are two key
points presented in each E-learning platform system. Our working interest reg-
isters in these two points, or the need for powerful tool to create automatic crea-
tion of course content and the source is of course the Web, which has a huge
space of information available requires good and over filtering. Our new tool
increase the quality of being given the wealth of Web resources, direct adapta-
tion based fusion of Web resources to learner profiles give high performance of
our new tool, and enrichment of courses directly from the Web with backup of
extracted resources, ensures the reusability of E-learning platform resources.
This is also important that teachers receiving full benefits, time and effort will
be reduced, and they just control over resources created in databases of system.
Keywords: Web Resources, E-learning Platform, Reusability, Adaptation, Fu-
sion, Learner Profiles.
1 Introduction
The amount of learning material on Internet has grown rapidly in recent decades.
Therefore, the information consumers are challenged to choose the right things. In
systems of e-learning, most approaches have led to confusion for learners. Inevitably,
adaptive learning has gained much attention in this area [1], [2].
We aim through our new tool reduce the huge space of the Web, containing bil-
lions of Web pages, to a personal space and direct adaptive to learners, to increase
their satisfaction and provide good training scalable to any change or update [3], but
with reliable and academic resources [4]. We must find good research and precise
filtering to extract the most relevant information, because we are facing a very large
mass of information available on the WEB, and editors spend an indefinite time to
create courses and more specifically, having a content database that will be adapted to
learner profiles. And before the learner needs to cultivate, to deepen more on such
field or theme of learning [5], we are obliged to produce system that uses the Web as
a documentary medium, and provides techniques to custom navigation for learners.
306
The rest of paper is organized as following: the second part related works and
learner needs to construct adaptive and personalized learning domain. In the third
part, we present our new tool and approach to create educational and adaptive courses
in an E-learning platform based fusion from Web resources. And finally, we terminate
with discussion and conclusion.
2 Related works
To create a practical learning environment for e-users, and to a broad audience (dif-
ferent objectives, knowledge levels, funds or learning abilities), it is necessarily that
the designers of e-learning systems thinking on adaptive learning environments and
flexible with this potential need, so they must improve the performance to the learners
[6], [7]. Recent works dealing with the problem of adaptation have very powerful
difficulty, because such learner profile can change a lot of time in learning [7], [3].
Some researchers are in making extensions for learning content standards to im-
prove the quality of learning process. These researchers argue that current standards
do not support an adaptive system so that they must be changed to have good adapta-
tion to learner model. Much effort has been made in the field of adaptive systems to
offer a user model. In learning systems, most of these works are about learning styles
of learners to gain more [8], [9], [10], [11]. Learning style is an acceptable factor of
adaptation, as it reflects the characteristics of learner preferences and needs.
There are two different general approaches of learning content adaptation [12],
[13], [14]. The first approach seeks to adapt learning content with special needs, and
the second focuses on the provision of the most appropriate learning content to needs
of learners. The first is called adaptation of content level and the second is called the
link-level adaptation. Neither approach has been preferred to another in the literature.
Several research projects have been targeted to lead to propose new methodology for
appropriate content. Some of these studies are underway on the extension of learning
content standards to improve the quality of learning process. One group argues that
current standards do not support an adaptive system so they must be modified in some
respects [15], [16]. In response to fact that metadata standards of learning content are
somehow inadequate for some applications, group of researchers tried to replace these
standards with ontologies 'Semantic Web' [17], [18], [19], [20], [21]. Ontologies
modeling course and give interaction between learners and systems, such as [17],
[18], [20], [21], [22]. There are some studies that have used agents in adaptive learn-
ing [3], [4].
Current generation of E-learning platform is not yet ready for commercialization
[23]. In other words, current studies are so focused on quality of adaptation [2] that
result in special systems designated for learning purposes and does not work with
other systems. In addition, no work to date has begun the next content before the ad-
aptation, that is, to adapt content unorganized or non-existent [24]. The new in our
research (addition to last work [25]), is the fusion of several fragments of Web re-
sources, to increase the quality of training content via adaptive, reliable, very rich and
dynamic learning domain in the sense enrichment and update.
307
3 Proposed approach
We must first searching in the Web by Google API; we can with this API finding Web
resources to be filtering in another processing step. In second time, we consider im-
plementation and the use of ontology in our system. A simple idea is the extraction of
concepts, slots and instances. To do this step, we need an API called Jena. This API
allows reading and writing of ontology (RDF or OWL type) in Java Platform. Our
domain ontology is OWL-type, which has facilitated its implementation. We keep the
hierarchy of ontology after extraction of concepts to give hierarchy that preparing our
Learning Domain to saving in next time all extracted segments in correspondent parts
of course in new segments database ‘NSDB’ after fusion of sub segments. ‘NSDB’
database use Excel model (as in Table 1.), to do this, we used a Java Excel API; this
API allows reading and writing an Excel document in Java Platform. For each part of
course, we define some semantic rules ‘SR’ to calculate degree of relevance ‘DR’ and
distance based semantic rules ‘DBSR’ of each sub segment ‘SS’ of one Web resource
part. The semantic rules of each course part defining in table are organized vertically
and for each one, we define their correspondent sub segments, these later are extracted
from Web resources. After this, we start fusion process (as in Fig. 1.) for each course
part in table, for example for Part 1, we choose the content stored in sub segment 1 to
N and save new segment or course part in correspondent column, in the same part of
course Part 1, we save result in FSS1 ‘Fusion of Sub Segments’.
Table 1. Portion of Excel Model to save Filtering Results
DR DBSR DR DBSR
Part 1 SR1 SS1 0 0 … SSN 0 0
FSS1 SR2 SS1 0 0 … SSN 0 0
… … 0 0 … … 0 0
SRN … 0 0 … … 0 0
… … … … … … … … …
… … … … … … … … …
Part N SR1 SS1 0 0 … SSN 0 0
FSSN SR2 SS1 0 0 … SSN 0 0
… … 0 0 … … 0 0
SRN … 0 0 … … 0 0
We obtained a comprehensive approach that meets our needs:
The hierarchy of course is mined from ontology of domain.
Annotations and keywords of each concept in ontology are extracted and assured
calculation of degree of relevance ‘DR’ (1) of each segment extracted from Web
resources to finding the most relevant portions.
We calculate Distance Based Semantic Rules ‘DBSR’ (2) for all relevant portions
to extract the most relevant sub segments.
Finally, we order the most relevant sub segments in Excel Model to create our New
Segments Database ‘NSDB’.
308
Fig. 1. Fusion process of three Web resources
3.1 Degree of Relevance ‘DR’
It is a statistical result (1), based on the frequency of ontology concept (which presents
a component of the course) in Web resource segment in the first part and the existence
of keywords and their frequencies in the same segment in the second part. The fre-
quency ‘F’ of a word in one segment is the number of times that word appears in this
segment. Degree of relevance equal frequency of ontology concept in fragment, adding
sum of keyword (k=0…n) frequencies of one ontology concept in one Web resource
segment, multiple by correspondent keyword weight ‘W’. The all is devised by total
number of words in one Web resource segment.
3.2 Distance Based Semantic Rules ‘DBSR’
It is a semantic result (2), based on the distance between terms in sub fragments, we
must firstly extract terms from one sub fragment, and we calculate distance only be-
tween terms that defined in semantic rules. DBSR present a projection of semantic
rule on sub fragment of Web resource to extract the most relevant sub segment appro-
priate to one sub part of course.
309
3.3 Fusion & Adaptation process
When processing of one document is finished, same steps were doing to other docu-
ments, but provided to relevant parts in order in Excel file for each component (col-
umn) of the course. If processing is completed, ‘NSDB’ database is full accomplished.
After that, our ‘NSDB’ database present mean of adaptation based fusion process. We
can adapt courses to level in learner profiles. Each level has number of course parts,
and number of semantic rules. If level augments, course parts and semantic rules aug-
ment.
4 Discussion and Conclusion
Through this study developed, we succeed in building new Web tool with new adap-
tation approach in E-learning platform, based research and filtering of Web resources,
after that, creating areas of learning with possibility of fusion of extracted resources,
and the most important, adaptation of Web content to learner profiles. The world in
the last years saw very rich side resources available on the Web; our method is to
reduce this informational space in an adaptive educational space, personalized and
mostly reusable for entire community of learners.
The study improves the quality of segments after fusion of several Web resources,
and reusability of segments stored in our database gives performance in E-learning
platform, and finally the augmentation of construction courses quality with enrichment
by Web resources and the good methods of research and filtering implicated in our
tool.
5 References
1. Chaoui, M., & Laskri, M-T.: Towards the Creation of Adaptive Content from Web Re-
sources in an E-Learning Platform to Learners Profiles. International Journal of World
Academy of Science, Engineering and Technology WASET. 77 (27), 157--162 (2011)
2. Caravantes, A., & Galn, R.: Generic Educational Knowledge Representation for Adaptive
and Cognitive Systems. Educational Technology & Society. 14 (3), 252--266 (2011)
3. Chen, C.: Intelligent Web-based learning system with personalized learning path guidance.
Computers & Education. 51(2), 787--814 (2008)
4. Canales, A., Pena, A., Peredo, R., Sossa, H., & Gutierrez, A.: Adaptive and intelligent
Web based education system: Towards and integral architecture and framework. Expert
Systems with Applications. 33(4), 1076--1089 (2007)
5. Papanikolaou, Mabbott, A., Bull, S., & Grigoriadou, M.: Designing learner-controlled ed-
ucational interactions based on learning/cognitive style and learner behavior. Interacting
with Computers. 18, 356--384 (2006)
6. Wang, M., Ran, W., Liao, J., and Yang, S.J.H.: A Performance-Oriented Approach to E-
Learning in the Workplace. Educational Technology & Society. 13(4), 167--179 (2010)
7. Chen, C.: Personalized E-Learning System with Self-Regulated Learning Assisted Mecha-
nisms for Promoting Learning Performance. Expert Systems with Applications. 36, 8816--
8829 (2009)
8. Yang, Y., & Wu, C.: An attribute-based ant colony system for adaptive learning objects
recommendation. Expert Systems with Applications. 36(2), 3034--3047 (2009)
310
9. Liegle, J., & Janicki, T.: The effect of learning styles on the navigation needs of Web-
based learners. Computers in Human Behavior. 22(5), 885--898 (2006)
10. Magoulas, G., Papanikolaou, K., & Grigoriadou, M.: Adaptive Web-based learning: Ac-
commodating individual differences through systems adaptation. British Journal of Educa-
tional Technology. 34(4), 511--527 (2003)
11. Stach, N., Cristea, A., & De Bra, P.: Authoring of learning styles in adaptive hypermedia.
In Proceedings of the 13th international World Wide Web conference on Alternate track
papers & posters. pp. 114--123 (2004)
12. Olfman, L., & Mandviwalla, M.: Conceptual versus procedural software training for
graphical user interfaces: A longitudinal field experiment. MIS Quarterly. 18(4), 405--426
(1994)
13. Papanikolaou, K., Grigoriadou, M., Magoulas, G., & Kornilakis, H.: Towards new forms
of knowledge communication: The adaptive dimension of a Web based learning environ-
ment. Computers & Education. 39(4), 333--360 (2002)
14. Samuelis, L.: Notes on the components for intelligent tutoring systems. Acta Polytechnica
Hungarica. 4(2), 77--85 (2007)
15. Lu, E., & Hsieh, C.: A relation metadata extension for SCORM content aggregation mod-
el. Computer Standards & Interfaces. 31(5), 1028--1035 (2008)
16. Rey-Lopez, M., Diaz-Redondo, R., Fernandez-Vilas, A., Pazos-Arias, J., Garcia-Duque, J.,
Gil-Solla, A., et al.: An extension to the ADL SCORM standard to support adaptivity: The
T-learning case study. Computer Standards and Interfaces. 31(2), 309--318 (2009)
17. Chi, Y.: Ontology-based curriculum content sequencing system with semantic rules, Ex-
pert Systems with Applications. 36(4), 7838--7847 (2009)
18. Jovanovic, J., Gasevic, D., Knight, C., & Richards, G.: Ontologies for effective use of con-
text in e-learning settings. Educational Technology & Society. 10(3). 47--59 (2007)
19. Lee, M., Tsai, K., & Wang, T.: A practical ontology query expansion algorithm for seman-
tic-aware learning objects retrieval. Computers & Education. 50(4), 1240--1257 (2008)
20. Shih, W., Yang, C., & Tseng, S.: Ontology-based content organization and retrieval for
SCORM-compliant teaching materials in data grids. Future Generation Computer Systems.
25(6), 687--694 (2009)
21. Zitko, B., Stankov, S., Rosic, M., & Grubisic, A.: Dynamic test generation over ontology-
based knowledge representation in authoring shell. Expert Systems with Application.
36(4), 8185--8196 (2009)
22. Lee, Y.-H., Hsieh, Y.-C., & Hsu, C.-N.: Adding Innovation Diffusion Theory to the Tech-
nology Acceptance Model: Supporting Employees' Intentions to use E-Learning Systems.
Educational Technology & Society. 14 (4), 124--137 (2011)
23. Whurle 2.0: An adaptive Web service e-learning environment for the Web
http://www.cs.nott.ac.uk/mzm/Saudi_Conference_mmeccawy13.pdf
24. Chaoui, M., & Laskri, M.T.: New method of finding information on the Web in unstruc-
tured information resources for educational use by learners, International Journal of Re-
search and Reviews in Computer Science, Science Academy Publisher, United Kingdom.
2 (1), 33--39 (2011)
25. Chaoui, M., & Laskri, M.T.: Automatic construction of an on-line learning domain. In
IEEE Proceedings of International Conference on Machine and Web Intelligence. pp. 439-
-443, November 2010.
311