Art. 05 – Vol. 21 – Nr. 4 – 2011

Sisteme adaptabile de e-Learning cu hărţi conceptuale

Rodica Potolea
Camelia Lemnaru
Florin Trif
Universitatea Tehnică din Cluj-Napoca

Rezumat:  Sistemele adaptabile de e-Learning reprezintă o nouă paradigmă în metodele moderne de învăţare. Ele nu se bazează doar pe asigurarea unei cantităţi mari de informaţii ci și pe calitatea transferului de cunoştinţe. În acest sens este esenţială identificarea corectă a stilului de învăţare al fiecărui utilizator pentru a i se oferi un conţinut adecvat. În plus, o reevaluare continuă şi o clasificare sunt importante pentru a face faţă progreselor realizate în timpul procesului de învăţare şi pentru a asigura o mai bună evoluţie. Hărțile conceptuale sunt un instrument important atât în dezvoltarea unui conţinut de înaltă calitate cât şi pentru evaluarea automată. Această lucrare prezintă un model pentru un sistem adaptabil de e-Learning şi detaliază modulele responsabile cu identificarea tipului de utilizator şi a hărţilor conceptuale.

Cuvinte cheie: sistem adaptabil de e-Learning, stil de învăţare, hărţi conceptuale, grupare, hărţi auto-organizate

Introducere: Following current trends triggered by the evolution of the technology, the education process has started to shift from the traditional face-to-face instruction to more modern approaches, such as online education, with the advantage of being available anywhere, anytime. The purpose of adaptive e-learning systems is to increase the student’s performance by adjusting the content and interaction methods to users with different interests, initial knowledge, background and skills. When confronted with the task of defining the user model, the developers of e-learning platforms rely on learning theories from educational psychology and pedagogy. There are some habitual ways of identifying user styles such us: answers to psychological tests and behavioral data observed from user interaction with the e-learning platforms. Current trends focus on the design of e-learning systems that contribute to the improvement the user’s performance during the learning process. The goal is no longer the acquisition of knowledge alone, but how to do it in the most appropriate manner for each individual.

Most of the online training systems are based on curricula segmentation, situation in which the students must go through a predefined structure. It is widely acknowledged that the student should be involved actively in the online learning process and that e-learning systems should sustain the student’s control and organization upon information [13]. Thus, the online training systems should constrain the user less, and should be able to adjust on his/her characteristic learning style. An essential element is anticipating the students’ behavior and adapting the content (both quantitatively and qualitatively) according to the needs. Thus, the structure of the courses and the segmentation of their presentation must be personalized according to the type of student.

An intelligent system should adjust the content in order to ensure faster learning and better performance. Moreover, it should help students develop new, desirable learning abilities.

This paper presents an efficient and accurate method for identifying the user typology in adaptive e-learning systems, and an original technique for automated evaluation of the knowledge acquired, using concept maps. The rest of the paper is organized as follows: section II presents the state of the art in adaptive e-learning systems; section III is a brief overview on the theory of learning styles, with focus on adaptive systems; section IV presents the general design of our proposed model for an adaptive e-learning system. Section V details the intelligent module, responsible for static and dynamic user type identification, while section VI presents the concept map paradigm, with focus on our method for automated evaluation of concept maps. The concluding remarks and future directions are presented in section VII.

Vizualizează articolul complet

Concluzii: We have presented our view on the adaptive e-learning strategy. We have designed, implemented and evaluated the model of an e-learning system containing elements from the aptitude-treatment strategy (via fixed factors measurements), the micro-adaptive approach (by measuring dynamic features via quick notes, navigation path and concept maps employment) and constructivist-collaborative approach (by means of the process coordinator). The most challenging task is represented by the intelligent module – the component that identifies the user’s type. Our current solution consists of a layered approach: a clustering layer, for the initial assessment, based on the fixed factors (user’s static features). The second layer consists of a SOM that receives both static and dynamic features. As the users’ interaction with the system intensifies, the structure of the SOM changes accordingly, indicating the current user type. The experiments on synthetic data have shown a correct identification of the four learning strategies mentioned in the literature. Moreover, they indicate an even better separation of the clusters by the evaluation of both static and dynamic features. This is a welcome validation of our assumption that dynamic attributes are better indicators of the evolution on the user learning style.

In terms of concept maps we have implemented a solution for automated CM evaluation: we proposed our own similarity evaluation, based on graph matching algorithms, and lexical and semantic content processing. The experiments performed so far showed that our solution is better in most cases than Cmap.

Our current interests focus on providing a thorough evaluation of our proposed model for user type identification, using various metrics, scenarios and a partition-based approach for the SOM, i.e. employ an identification-based strategy rather than a separation-based one. Thus, one SOM for each user learning style can be separately built and tuned, resulting in a more accurate model for each type.


This work has been supported by a grant from Romanian Ministry of Education, Research and Innovation, CNMP, grant no. 12-080/ 1.10.2008, SEArCH – Adaptive E-learning Systems using Concept Maps.


  1. CAÑAS, A. J.; REISKA, P.; ÅHLBERG, M.; NOVAK, J. D. (EDS.): Concept Mapping: Connecting Educators Proc. of the Third Int. Conference on Concept Mapping Tallinn, Estonia & Helsinki, Finland 2008.
  2. CHANG, Y.C.; KAO, W.Y.; CHU, C. P.; CHIU, C. H.: A learning style classification mechanism for e-learning. Computers & Education, article in press, journal homepage: 2009.
  3. CHEN, C. M.; LEE, H. M.; CHEN, Y. H.: Personalized e-learning system using Item Response Theory. Computers and Education, 44(3), 2005, pp. 237-255.
  4. FIRTE, A. A.; VIDRIGHIN BRATU, C.; CENAN, C.: Intelligent component for adaptive E-learning systems. Proceedings of the IEEE 5th International Conference on Intelligent Computer Communication and Processing, ICCP 2009, 27-29 August 2009, Cluj-Napoca, Romania, 2009, pp. 35-38.
  5. LEMNARU, C.; FIRTE, A. A.; POTOLEA, R.: Static and Dynamic User Type Identification in Adaptive E-learning with Unsupervised Methods. Proceedings of ICCP 2011, 2011, pp.11-18.
  6. FIRTE, A.A.: Metode Nesupervizate de Identificare a Sabloanelor in Date. Identificarea Tipologiei Utilizator in sisteme e-Learning Adaptive. Master thesis, Technical University of Cluj-Napoca, July 2011.
  7. GALE, D.; SHAPLEY, L. S.: College Admissions and the Stability of Marriage. American Mathematical Monthly 69, 1962, pp. 9-14.
  8. GARCÍA, P.; AMANDI, A.; SCHIAFFINO, S.; CAMPO, M.: Evaluating Bayesian Networks’ precision for detecting students’ learning styles. Computers and Education, 49(3), 2007, pp. 794-808.
  9. HLAOUI, A.; WANG, S.: Image Retrieval Systems Using Graph Matching. Rapport de Recherche, No. 275, Département de mathématiques et d’informatique, Univérsité de Sherbrooke, 2001.
  10. KRITIKOU, Y.; DEMESTICHAS, P.; ADAMOPOULOU, E.; DEMESTICHAS, K.; THEOLOGOU, M.; PARADIA, M.: User Profile Modeling in the context of web-based learning management systems. Journal Network. Comput. Appl.31, 4 (Nov. 2008), 2008, pp. 603-627.
  11. KUHN, HAROLD W.: Variants of the Hungarian method for assignment problems. Naval Research Logistics Quarterly, 3: 253-258, 1956.
  12. MERA, V.: Compararea Automata a Hartilor Conceptuale. Evaluarea in Sisteme e-Learning Adaptive. Master thesis, Technical University of Cluj-Napoca, July 2011.
  13. MITCHEL, T.; CHEN, S.Y.; MACREDIE, R.: Adapting hypermedia to cognitive Styles: it is necessary? Individual differences in Adaptive Hypermedia. Proceedings of the AH Workshop, 2004, pp. 70-79.
  14. MÖDRITSCHER, F.: Adaptive E-Learning Environments: Theory, Practice, and Experience. VDM Verlag, 2008.
  15. NOVAK, J. D.; CAÑAS, A. J.: The Theory Underlying Concept Maps and How to Construct Them. Technical Report IHMC CmapTools 2006-01 Rev 01-2008, Florida Institute for Human and Machine Cognition, 2008, available at: ResearchPapers/TheoryUnderlyingConceptMaps.pdf
  16. PAPANIKOLAOU, K. A.; GRIGORIADOU, M.: Accommodating learning styles characteristics in Adaptive Educational Hypermedia Systems. Individual differences in Adaptive Hypermedia, Proceedings of the AH Workshop, 2004, pp. 1-11.
  17. TRIF, F.; LEMNARU, C.; POTOLEA, R.: Identifying the User Typology for Adaptive E-learning Systems. Proceedings of  AQTR 2010 – IEEE International Conference on Automation, Quality and Testing, Robotics 2010, Tome III, 2010, pp. 192-198.
  18. SCHIAFFINO, S.; GARCIA, P.; AMANDI, A.: eTeacher: Providing personalized assistance to e-learning students. Computers & Education, 51, 2008, pp. 1744-1754.
  19. WITKIN, H. A.; MOORE, C. A.; GOODENOUGH, D. R.; COX, W.: Field dependent and field independent cognitive styles and their educational implications. Review of Educational Research, 47, 1977, pp. 1-64.
  20. WOOLF, B.P.: Building intelligent interactive tutors. Burlington, Elsevier: Morgan Kaufman Publisher, 2009, pp. 44-45

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.