Smart Collaborative Learning: A Recommended Building Team Approach

Smart Collaborative Learning: A Recommended Building Team Approach

Ouidad Akhrif, Chaymae Benfares, Younès El Bouzekri El Idrissi, Nabil Hmina
Copyright: © 2019 |Pages: 15
DOI: 10.4018/IJSST.2019070103
Article PDF Download
Open access articles are freely available for download


Technologically enhanced learning has shifted from digital resources to smart components to afford more content, support tools, and provide learning guidance that meets a learner's needs and interests by delivering smart university services. Smart interaction is essential in smart university, it is a concept that offers new opportunities and new channels of communication between learners. This communication is reinforced by the concept of collaboration, an important factor for knowledge sharing. The current study concerns team building based on the recommendation of the most appropriate collaborator in order to make groups of learners promoting universal participation of all members of the team. The complexity of this problem requires collaborative filtering algorithms to find the potential collaborators for each learner, taking into account problem-solving as a parameter representing items of the recommendation matrix.
Article Preview


The appearance of innovative technology-based learning and teaching strategies has involved the educational government. Additionally, technology-based learning allows moderns learning approaches, that encourages students to be more interactive and engaged within a team of workers. Smart collaborative pedagogy (SCP) guides educators and students to use technology for collaboration and navigation of the potentially conflicting role of autonomous collaborative learning. Furthermore, SCP highlights the importance of students contributing personal meanings and using appropriate communication strategies as they work together using interactive technologies in innovative ways.

This approach supports the use of collaboration in an academic environment and more specifically in a smart university (SU), which offers opportunities for smart interactions. Smart interactions are mechanisms of transmission and technological means through which the learner interacts with their environment, promoting his participation, collaboration, and optimization of their capabilities. In fact, the modernization of learning techniques has emerged a new way of interacting between different stakeholders of an SU. This new generation of interaction requires recent data processing techniques, in order to offer smart services adapted to a learner profile, in terms of accessibilities and capabilities. Among these techniques, recommendation system remains an essential concept in the implementation of these approaches, through which the student benefits from ample services tailored according to their needs, performances, and competences. In addition, this creates a collaborative workspace that allows the sharing and acquisition of knowledge in an optimal, efficient, and intelligent way.

In an academic environment, team building plays a key role in the acquisition of knowledge and skills through courses and practical work. Through team building, the learner composition has a specificity compared to what can be found in a professional environment in terms of objectives, participants, and means. In fact, the university’s collaboration aims at the universal integration of all learners into working groups and the ability to share information in a fair way. This complexity has led the authors to think of a method for building learning teams by promoting universal and participatory integration that is mainly based on a collaborative filtering algorithm.

In this study, the authors investigate the introduction of a recommendation system in the SU in order to create work teams characterized by their complementarity and efficiency through the selection of the most appropriate collaborator in a work team, and the assignment of tasks to the learner by taking into account its accessibility and capabilities. To do this, the authors have used a memory-based collaborative filtering recommendation system, which better responds to this challenge, basing on three steps: 1) Neighborhood identification; 2) Predicting calculation; and 3) Active collaborator selection. This allows the student to benefit from many services adapted to their profile and ensures interaction between the different stakeholders of an SU.

The remaining parts of the paper are mainly structured in sections that include a review of the literature focused on background information on the analysis of the present situation, as explained in Section 2. Section 3 consists of explanations regarding the proposed methods of collaborative learning (CL), followed by the system architecture, functional algorithms, and effective parameters. Section7 explains the validation process and Section 8 discusses the conclusion.

Complete Article List

Search this Journal:
Volume 10: 1 Issue (2024)
Volume 9: 2 Issues (2022): 1 Released, 1 Forthcoming
Volume 8: 2 Issues (2021)
Volume 7: 2 Issues (2020)
Volume 6: 2 Issues (2019)
View Complete Journal Contents Listing