Content uploaded by Riichiro Mizoguchi
Author content
All content in this area was uploaded by Riichiro Mizoguchi on May 10, 2020
Content may be subject to copyright.
A preview of the PDF is not available
How Can We Form Effective Collaborative Learning Groups?
Abstract
Our research objectives include constructing a collaborative learning support system that detects appropriate situation for
a learner to join in a collaborative learning session, and forms a collaborative learning group appropriate for the situation
dynamically. In this paper, we describe a system of concepts concerning learning goals expected to attain by learners through
collaborative learning process with justification by the learning theories. With the ontology, it will be possible to compare
and synthesize the learning theories to design the collaborative learning settings.
... Group formation is an NP-Hard problem [8]. Previous research has studied various methods for group formation, such as rule/inference [9], Multi-agent [10] [11], Greedy algorithm [12], Genetic Algorithm [13], Hill Climbing [14], Fuzzy C-Means [15], Ant Colony Optimization [16] , and Semantic Web [8]. ...
... So that, it can help teachers to develop group formation system based on their cases. The various criteria or attributes have been applied in group formation, such as knowledge or expertise in a specific domain [9][15], learning goal [10], learners' performance in previous teamwork [10] [16], personality traits [16], learning style [8][17] [18], thinking style [13], Belbin role and minority [8], and preferred time slots and project [12]. ...
... So that, it can help teachers to develop group formation system based on their cases. The various criteria or attributes have been applied in group formation, such as knowledge or expertise in a specific domain [9][15], learning goal [10], learners' performance in previous teamwork [10] [16], personality traits [16], learning style [8][17] [18], thinking style [13], Belbin role and minority [8], and preferred time slots and project [12]. ...
... In collaborative learning (CL), group formation poses as an important and complex element for designing successful CSCL scenarios (Inaba et al., 2000). According to Dillenbourg (2002), forming groups without any criteria or strategy (e.g., randomly) is usually ineffective, since there is no guarantee that such composition will trigger the expected interactions and learning among their members. ...
... In the CL context, Inaba et al. (2000) was one of the first studies to use ontologies for describing CL scenarios. Based on that, other studies were conducted for the creation of models and strategies to support, for example, the planning of collaborative activities, the formation of learning groups, and the engagement/interaction among students (Inaba & Mizoguchi, 2004;Isotani et al., 2009;Isotani et al., 2013;Challco et al., 2015;Reis et al., 2016). ...
Group formation is an important and challenging element for designing successful CSCL scenarios. Despite efforts from the scientific community in developing more effective algorithms to support group formation processes, we still face problems related to learners’ resistance and demotivation towards group work. In this sense, diverse studies highlight the importance of considering learners’ personality traits to form groups, since this factor can influence students’ performance and induce diverse actions and behaviors in group work. Therefore, this paper presents G-FusionPT (Group Formation USIng Ontology and Personality Trait), a group formation algorithm that support new learning roles, denominated Affective Collaborative Learning roles, based on relation between collaborative learning theories and students’ personality traits. The algorithm is based on a collaborative ontology to understand the learning theories (e.g., context, learning activities, group structure), and learners profile to understand learners’ needs (e.g., target/current knowledge/skill). To evaluate the algorithm, we used a 300 student simulated sample wit varying group size (three, five, and seven members), and compared G-FusionPT results to other group formation algorithms: G-Fusion (based specifically on collaborative learning theories) and Random (no strategy or criterion). The results demonstrated the effectiveness of G-FusionPT against G-Fusion and Random algorithms, as it generated the highest average percentage of learners in well-formed groups and lowest averagepercentage of learners in unfit groups.
... The composition of student groups impacts their ability to collaborate because not all groups have the same ability to collaborate. For example, researchers (Cress and Kimmerle 2008;Inaba et al. 2000;Issroff and Jones 2005) found that various student attributes such as their knowledge and collaboration skills impact their collaborative learning outcomes. So, a Wiki environment that models the students and their collaborations to learn the composition of members that improves their collaboration is useful. ...
Wikis are being increasingly used as a tool for conducting colla-borative writing assignments in today’s classrooms. However, Wikis in general (1) do not provide group formation methods to more specifically facilitate collaborative learning of the students and (2) suffer from typical problems of collaborative learning like detection of free-riding (earning credit without contribution). To improve the state of the art of the use of Wikis as a collaborative writing tool, we have designed and implemented ClassroomWiki - a Web-based collaborative Wiki that utilizes a set of learner pedagogy theories to provide multiagent-based tracking, modeling, and group formation functionalities. For the students, ClassroomWiki provides a Web interface for writing and revising their group’s Wiki and a topic-based forum for discussing their ideas during collaboration. When the students collaborate, ClassroomWiki’s agents track all student activities to learn a model of the students and use a Bayesian Network to learn a probabilistic mapping that describes the ability of a group of students with a specific set of models to work together. For the teacher, Clas-sroomWiki provides a framework that uses the learned student models and the mapping to form student groups to improve the collaborative learning of students. ClassroomWiki was deployed in three university-level courses and the results suggest that ClassroomWiki can (1) form better student groups that improve stu-dent learning and collaboration and (2) alleviate free-riding and allow the instructor to provide scaffolding by its multiagent-based tracking and modeling.
... Machine learning approaches for evaluating large amounts of educational data can provide vital information, potentially with large impacts on future education (Bienkowski et al., 2012). For example, highly accurate machine learning models have been constructed to predict the time students are required to generate a response and to estimate the likelihood that the student's response was correct (Inaba et al., 2000). Saarela et al. (2016) used a combination of unsupervised and supervised learning algorithms to predict student performance on math scores, which is a unique way of learning directly from large-scale educational assessment studies' (LSAs) data. ...
... Pour enrichir la classification des tumeurs hépatiques détectées dans les images IRM, nous avons intégré l'aspect sémantique au processus d'apprentissage. Ces informations sont extraites à partir des ontologies OntHCC [9] et MROnt [10]. L'OntHCC, présentée dans le chapitre précédent, (voir Figure IV.4, ...
Le diagnostic des lésions hépatiques est une tâche complexe surtout lorsque les nodules détectés sont de petites tailles. Dans ce cas, il devient très difficile de connaitre leurs natures (tumeur bénigne ou maligne, type de lésion, etc). Dans des cas similaires, il faut répéter des examens cliniques pendant plusieurs mois pour voir l’évolution des masses hépatiques. Afin de mieux répondre à ces problèmes, il faut trouver des solutions informatiques qui servent à l’optimisation du diagnostic des tumeurs du foie. Dans le contexte de la classification des lésions hépatiques, nous avons développé une première approche ontologique (OntHCC) pour l’aide au diagnostic, à la stadification et au choix de traitement des tumeurs CHC (Carcinome Hépatocellulaire). Cette approche est fondée sur l’analyse d’images IRM de foies infectés et sur des rapports radiologiques. Par la suite, nous avons proposé une deuxième approche ontologique (MROnt) pour la modélisation de l’information médicale contenue dans les rapports radiologiques, dans le cadre du diagnostic et de suivi de tumeurs du foie. La détection automatique des tumeurs du foie nécessite un processus de diagnostic primaire en utilisant obligatoirement les images médicales (par exemple IRM ou scanner). Pour ce faire, nous avons intégré l’apprentissage profond dans la classification d’images IRM avec prise de contraste. Dans la suite de la thèse et afin d’accroitre la performance du processus de classification des images, nous avons intégré les connaissances sémantiques. L’objectif est de profiter de la base de connaissances offerte par les ontologies pour décrire les images médicales et fournir des informations sur les tumeurs détectées (par exemple, le type, la taille et le stade). En outre, notre approche consiste à développer un CNN multi-label afin de supporter les ontologies développées (OntHCC et MROnt). Nous montrons l’efficacité des approches et prototypes proposés dans ces travaux de thèse à travers des évaluations numériques comparatives et des études de cas.
... However, there has been relatively little attention given to this problem by the VLA community. Some studies such as [11,12] have looked at the relevance of forming groups based on an ontological description of the learning goals and collaborative environment, but in our study, we grouped students based on their similarity levels and activity levels. Our goal is to develop a grouping tool that supports teachers in finding relevant groups of students with similar learning outcomes and activities that are likely to result in meaningful decision-making to foment students' collaborative learning. ...
With the help of Visual Learning Analytics (VLA) tools, teachers can construct meaningful groups of students that can, for example, collaborate and be engaged in productive discussions. However, finding similar samples in large educational databases requires effective similarity measures that capture the teacher’s intent. In this paper we propose a web-based VLA tool called Similarity-Based Grouping (SBGTool), to assist teachers in categorizing students into different groups based on their similar learning outcomes and activities. By using SBGTool, teachers may compare individual students by considering the number of answers (correct and incorrect) in different question categories and time ranges, find the most difficult question categories considering the percentage of similarity to the correct answers, determine the degree of similarity and dissimilarity across students, and find the relationship between students’ activity and success. To demonstrate the tool’s efficacy, we used 10,000 random samples from the EdNet dataset, a large-scale hierarchical educational dataset consisting of student-system interactions from multiple platforms, at university level, collected over a period of two years. The results point to the conclusion that the tool is efficient, can be adapted to different learning domains, and has the potential to assist teachers in maximizing the collaborative learning potential in their classrooms.
... These ideas were taken up and extended in I-Help and PHELPS (Greer et al. 1998) in large-scale practical applications. Also in the context of AIED research, Inaba et al. (2000) have proposed an ontology-based approach to group formation that explicitly represents assumptions from underlying learning theories that are formalized and encoded in the ontology. ...
Collaborative learning in small groups can enrich and enhance the learning experience in large online courses by facilitating interaction and collaborative knowledge building between peers. on the work reported here addresses scenarios based on asynchronous communication and exchange. As compared to synchronous and face-to-face settings, these scenarios require higher explicit coordination efforts and specific support mechanisms to reach out to inactive group members. This can be achieved by human tutors as well as through automatic interventions by the system in the organization and facilitation of group work. Our approach is based on characterizing the quality or well-functioning of collaboration using sequence alignment techniques that were originally developed in bioinformatics for the comparison of DNA sequences. Sequence alignment provides a similarity measure between pairs of action logs originating from the group work. The ensuing similarity matrix allows for clustering the working groups characterized by the similarity of their complete action logs. Notably, these clusters differ clearly in terms of quality measures related to the activity distribution and group output. This allows for determining quality indicators that can trigger feedback and adaptive scaffolding. Our data stem from inter-university online courses in which regular (weekly or bi-weekly) assignments had to be completed in small groups supported by an online forum for coordination and a separate tool for the actual collaborative writing. The basic actions are interpreted/coded as different contribution types, namely coordination, monitoring, minor/major contributions. Longer periods of inactivity are coded as another type of action (“gap”). These are the building blocks for the sequence alignment. One consequence of our analysis is a revision of the intuitive assumption that inactivity is a main indicator of inefficient group work: We found that inactivity can be compensated by early coordination to the extent that coordination is even the most important factor for successful group work. These insights are now being used to design alert mechanisms to signal inefficient group work and to generate timely interventions.
Intelligent Face Tracking for Collaborative Synchronous e-Learning using Pattern Recognition Model can provide high levels of interaction for distance learning initiatives. With the rapid evolution of technology, face recognition login and tracking, continuous product evaluation is necessary to ensure optimal methods and resources for connecting students, instructors, and educational content in rich, online learning communities. This article presents the analysis of online, synchronous learning solutions. Focusing on their abilities to meet technical and pedagogical needs in higher education. To make a solid comparison, the systems were examined in online classrooms with instructors, guest speakers, and students. Relative to usability, instructional needs, technical aspects and compatibility are outlined for systems.
University bilingual science courses are on the rise because of the internationalisation of higher education. These courses are legitimate and desirable scenarios for the promotion of bilingual written scientific argumentation (BWSA) as part of bilingual scientific literacy. The problem is that very little evidence is available on how to foster university student BWSA in courses where the English (foreign language) level of the undergraduates is variable. The goal of this study was to provide evidence that collaborative learning (CL) can be used to promote BWSA in contexts where students with different levels of English proficiency study together. We examined the Spanish-English bilingual written argumentation produced by 56 undergraduates (30 females and 26 males, 16–30 years old) in Colombia during a university Spanish-English bilingual science course where the English (foreign language) level of the students was variable. The data used in this analysis was derived from students’ written responses and audio recordings. The findings show that CL can contribute to BWSA as well as to specific CL skills (e.g. interdependence, individual accountability, interaction) which are important for social life. Educational implications for higher education are discussed.
