Modelling collaborative problem-solving competence with transparent learning analytics

Cukurova, Mutlu; Zhou, Qi; Landolfi, Lorenzo

doi:10.1145/3375462.3375484

Cited by 31 publications

(28 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(D) Can we make Learning Analytics solutions transparent to the end-users? Explainable Learning Analytics (XLA) focuses on enhancing the transparency of Machine Learning techniques in support of the Learning Analytics domain [33]. Since most of the ML models act as black-boxes, the outcomes are often hardly interpretable.…”

Section: Literature Reviewmentioning

confidence: 99%

Predicting Student Academic Performance by Means of Associative Classification

et al. 2021

View full text Add to dashboard Cite

The Learning Analytics community has recently paid particular attention to early predict learners’ performance. An established approach entails training classification models from past learner-related data in order to predict the exam success rate of a student well before the end of the course. Early predictions allow teachers to put in place targeted actions, e.g., supporting at-risk students to avoid exam failures or course dropouts. Although several machine learning and data mining solutions have been proposed to learn accurate predictors from past data, the interpretability and explainability of the best performing models is often limited. Therefore, in most cases, the reasons behind classifiers’ decisions remain unclear. This paper proposes an Explainable Learning Analytics solution to analyze learner-generated data acquired by our technical university, which relies on a blended learning model. It adopts classification techniques to early predict the success rate of about 5000 students who were enrolled in the first year courses of our university. It proposes to apply associative classifiers at different time points and to explore the characteristics of the models that led to assign pass or fail success rates. Thanks to their inherent interpretability, associative models can be manually explored by domain experts with the twofold aim at validating classifier outcomes through local rule-based explanations and identifying at-risk/successful student profiles by interpreting the global rule-based model. The results of an in-depth empirical evaluation demonstrate that associative models (i) perform as good as the best performing classification models, and (ii) give relevant insights into the per-student success rate assignments.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Predicting Student Academic Performance by Means of Associative Classification

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The orientation towards products or outcomes is also evident across other research on collaborative problem solving (Hmelo-Silver & Barrows, 2008;Praharaj et al, 2018). In fact, many prior research studies have the objective of improving student learning or helping them produce better artifacts (Schneider et al, 2018;Worsley & Blikstein, 2018;Spikol, Superiore, Anna, & Landolfi, 2020;Martinez-Maldonado et al, 2017), but are not necessarily concerned with team dynamics or how much each person is contributing.…”

Section: Constructivementioning

confidence: 97%

Designing Analytics for Collaboration Literacy and Student Empowerment

Worsley¹,

Anderson²,

Melo³

et al. 2021

JLA

View full text Add to dashboard Cite

Collaboration has garnered global attention as an important skill for the 21st century. While researchers have been doing work on collaboration for nearly a century, many of the questions that the field is investigating overlook the need for students to learn how to read and respond to different collaborative settings. Existing research focuses on chronicling the various factors that predict the effectiveness of a collaborative experience, or on changing user behaviour in the moment. These are worthwhile research endeavours for developing our theoretical understanding of collaboration. However, there is also a need to centre student perceptions and experiences with collaboration as an important area of inquiry. Based on a survey of 131 university students, we find that student collaboration-related concerns can be represented across seven different categories or dimensions: Climate, Compatibility, Communication, Conflict, Context, Contribution, and Constructive. These categories extend prior research on collaboration and can help the field ensure that future collaboration analytics tools are designed to support the ways that students think about and utilize collaboration. Finally, we describe our instantiation of many of these dimensions in our collaborative analytics tool, BLINC, and suggest that these seven dimensions can be instructive for re-orienting the Multimodal Learning Analytics (MMLA) and collaboration analytics communities.

show abstract

“…Due to their low financial and technical costs, video-based analytics of collaborative learning can provide valuable opportunities for immediate real-world impact. Although, there is early work investigating the potential of video data to analyze learner behaviors in collaborative learning activities through modelling learner behaviours [7], there is also a large need for developing novel computational approaches. For instance, in [7], CPS behaviours were modelled using a traditional classification approach (decision trees) and a semiautomated pipeline (active, semi-active, passive engagement values were manually coded to model CPS competence).…”

Section: Related Workmentioning

confidence: 99%

“…Although, there is early work investigating the potential of video data to analyze learner behaviors in collaborative learning activities through modelling learner behaviours [7], there is also a large need for developing novel computational approaches. For instance, in [7], CPS behaviours were modelled using a traditional classification approach (decision trees) and a semiautomated pipeline (active, semi-active, passive engagement values were manually coded to model CPS competence). The authors prioritised the transparency of the models over their performance and presented their results as an opportunity for humans to better interpret the models.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Inferring Student Engagement in Collaborative Problem Solving from Visual Cues

Kasparova

Çeliktutan

Cukurova

2020

Companion Publication of the 2020 International Conference on Multimodal Interaction

Self Cite

View full text Add to dashboard Cite

Automatic analysis of students' collaborative interactions in physical settings is an emerging problem with a wide range of applications in education. However, this problem has been proven to be challenging due to the complex, interdependent and dynamic nature of student interactions in real-world contexts. In this paper, we propose a novel framework for the classification of student engagement in open-ended, face-to-face collaborative problem-solving (CPS) tasks purely from video data. Our framework i) estimates body pose from the recordings of student interactions; ii) combines face recognition with a Bayesian model to identify and track students with a high accuracy; and iii) classifies student engagement leveraging a Team Long Short-Term Memory (Team LSTM) neural network model. This novel approach allows the LSTMs to capture dependencies among individual students in their collaborative interactions. Our results show that the Team LSTM significantly improves the performance as compared to the baseline method that takes individual student trajectories into account independently. CCS CONCEPTS• Computing methodologies → Computer vision tasks; • Applied computing → Interactive learning environments.

show abstract

Modelling collaborative problem-solving competence with transparent learning analytics

Cited by 31 publications

References 25 publications

Predicting Student Academic Performance by Means of Associative Classification

Predicting Student Academic Performance by Means of Associative Classification

Designing Analytics for Collaboration Literacy and Student Empowerment

Inferring Student Engagement in Collaborative Problem Solving from Visual Cues

Contact Info

Product

Resources

About