Volume 24-Issue1-2015-TUDOR

Liviana TUDOR¹, Adrian MOISE^2
1 Dept. of Informatics, Information Technology, Mathematics and Physics,
Petroleum – Gas University of Ploiesti
39, Bd. Bucuresti, 100680, Ploiesti, Romania
LTudor@upg-ploiesti.ro
² Dept. of Automatic Control, Computers and Electronics,
Petroleum – Gas University of Ploiesti
Bd. Bucuresti Nr. 39, 100680, Ploiesti, Romania
AMoise@upg-ploiesti.ro

Abstract: In this paper, the authors describe an open and collaborative model based on metacognitive neural strategies applied to students’ research projects in Artificial Neural Networks. The model is focused both on instructor–student and student–student interactions and includes an on-line knowledge-based intelligent system as well as several social psychological determinants. The performance of the proposed model is proved by the results obtained by the MSc students, their motivation to complete a postgraduate course and it is justified by the highly useful online knowledge base. The functionality and efficiency of the intelligent system are highlighted by traffic and log analysis of the Open Source data collections used in the disciplines which have been involved in the experiment.

Keywords: open and collaborative learning model, metacognitive strategies, artificial neural networks, online knowledge based intelligent system, dissertation works.

>Full text
CITE THIS PAPER AS:
Liviana TUDOR, Adrian MOISE, Open and Collaborative Learning Model Based on Metacognitive Strategies, Studies in Informatics and Control, ISSN 1220-1766, vol. 24 (1), pp. 71-78, 2015.

1. Introduction

The significant progress in Information Technology in the last years has imposed a steady training process of students in Artificial Intelligence and has also required a certain adaptation of the university curriculum in order to integrate knowledge about automatic learning and machine learning [1]. Automatic learning studies systems that are able to improve their performance based on a set of training data [2].

The academic curriculum for MSc studies in Informatics and Systems Engineering could be a component of the student-centred curricula. A curriculum is a set of teaching, learning and evaluation competences that allows the university to offer the students a whole system of information, skills, behaviours and competencies. One of the optional subjects in a university, for Informatics MSc level, allows the development of skills in the Artificial Neural Networks (ANN) field and belongs to the Artificial Intelligence curricula area. The status of the ANN as an elective field of study means it could be studied, at the students’ request, bringing their needs and interests to the foreground.

ANN teaching and learning should be based on an Open and Collaborative Learning Model (OCLM) and on interactive teaching methods. Collaborative learning of neural algorithms has as a major goal the training of students so they can acquire interdisciplinary skills and carry out team research projects, in a cooperation and competition framework. The instructor – student collaboration is based on developing and using neural metacognitive strategies. The metacognitive action, the thought about thought, represents a high-level cognitive activity. When it is related to the learning activity, the metacognitive competence defines a measure of: what a student knows, what a student knows he doesn’t know, the knowledge and perception about how learning occurs and the manner to make the learning process efficient [3].

The original contribution of this paper lies in the methodological and scientific aspects and is summarized below. (1) A collaborative teaching and learning method, based on scientific research projects in ANN, will be described. The impact on the efficiency of teaching activity will be studied, while using the collaborative teaching and learning method based on neural metacognitive competencies. An experimental study makes an analysis of the MSc students results over two years of study. (2) Development of an Open and Collaborative Learning Model (OCLM) can be done by building a virtual teaching, learning and evaluation environment; that means, a software package that combines Web technologies with Artificial Intelligence techniques applied to education. The adaptation process of the OCLM to the MSc students’ expectations is done by developing intelligent systems based on online knowledge.

The paper is structured as follows. In the section titled Related works the authors present the main research results in using Artificial Intelligence