Collaborative learning tools play important roles in communications and knowledge building, among learners in a virtual learning environment. They demand appropriate grouping algorithms as well as facilitating learners’ participations mechanisms. This paper has utilized More
Collaborative learning tools play important roles in communications and knowledge building, among learners in a virtual learning environment. They demand appropriate grouping algorithms as well as facilitating learners’ participations mechanisms. This paper has utilized some information retrieval techniquesto investigate the relevance of discussion posts to their containing forums, and extract learners’ most frequent topics. Trying to explore students online interactions, researchers have applied social network analysis, which has led to a new representation of social networking. They have proposed a new grouping algorithm based on the provided representation of social relationships. The mentioned approaches have been evaluated in some academic courses in Department of Electrical and Computer Engineering, and ELearning Center, University of Tehran. The results have revealed some considerable improvements in comparison to the traditional approaches. Research outcomes may help tutors to create and guide groups of learners more effectively.
Manuscript profile
In the case of penetration maximization, the goal is to find the minimum number of nodes that have the most propagation and penetration in the network. Studies on maximizing penetration and dissemination are becoming more widespread. In recent years, many algorithms hav More
In the case of penetration maximization, the goal is to find the minimum number of nodes that have the most propagation and penetration in the network. Studies on maximizing penetration and dissemination are becoming more widespread. In recent years, many algorithms have been proposed to maximize the penetration of social networks. These studies include viral marketing, spreading rumors, innovating and spreading epidemics, and so on. Each of the previous studies has shortcomings in finding suitable nodes or high time complexity. In this article, we present a new method called ICIM-GREEDY to solve the problem of maximizing penetration. In the ICIM-GREEDY algorithm, we consider two important criteria that have not been considered in the previous work, one is penetration power and the other is penetration sensitivity. These two criteria are always present in human social life. The proposed method is evaluated on standard datasets. The obtained results show that this method has a better quality in finding penetrating nodes in 30 seed nodes than other compared algorithms. This method also performs better in terms of time compared to the comparative algorithms in terms of relatively fast convergence.
Manuscript profile
Today, thanks to the growth and development of communication and information technologies, online learning systems have been able to provide group learning facilities and space for interaction and exchange of ideas between learners. This requires the formation of effect More
Today, thanks to the growth and development of communication and information technologies, online learning systems have been able to provide group learning facilities and space for interaction and exchange of ideas between learners. This requires the formation of effective learning groups and the provision of learner participation tools in online learning environments, which is rarely seen in existing systems that use virtual learning centers. In this article, the content and structure of the discussion forums have been examined. Content analysis has been done in order to match the content of the discussions with the objectives of the forum and to extract the areas of interest of the participants. While expressing the achievements of the social network analysis of an academic learning environment, the researchers have presented a solution for extracting the social relationships of people through the structural analysis of discussion forums in an online learning environment. Also, they have presented a method to use the extracted relationships in the mechanisms of grouping learners and evaluated its efficiency. Different parts of this research have been conducted in different courses in consecutive semesters and its achievements can be used to improve collaborative learning activities in online and blended learning environments.
Manuscript profile
Social network analysis’ link prediction has a diverse range of applications in different areas of science. Bipartite networks are a kind of complex network, which can be used to describe various real-world phenomena. In this article, a link prediction method for bipart More
Social network analysis’ link prediction has a diverse range of applications in different areas of science. Bipartite networks are a kind of complex network, which can be used to describe various real-world phenomena. In this article, a link prediction method for bipartite network is presented. Uni-partite link prediction methods are not effective and efficient enough to be applied to bipartite networks. Thus, to solve this problem, distinct methods specifically designed for bipartite networks are required. The proposed method is neighbor based and consisted of measures of such. Classic uni-partite link prediction measures are redefined to be compatible with bipartite network. Subsequently, these modified measures are used as the basis of the presented method, which in addition to simplicity, has high performance rates and is superior to other neighbor-based methods by 15% in average.
Manuscript profile
