Student Engagement Dataset

Origgi, Juan Manuel; Hasanpoor, Tania; Yu, Haiming; Allessio, Danielle; Arroyo, Ivon; Lee, William; Betke, Margrit; Woolf, Beverly Park; Bargal, Sarah Adel

doi:10.1109/iccvw54120.2021.00405

Cited by 16 publications

(5 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In the Learning Analytics research community, several open datasets are available that contain face or log data, which are commonly used for various purposes. For example, the DAISEE dataset (Gupta et al, 2022 ) provides facial video data with annotations of learning engagement, while the student engagement dataset (Delgado et al, 2021 ) offers in-the-wild facial video data for learning engagement analysis. Additionally, the ASSISTment dataset 1 provides response log data for educational assessment.…”

Section: Methodsmentioning

confidence: 99%

State-Aware Deep Item Response Theory using student facial features

Zhou,

Suzuki,

Kumano

2024

Front. Artif. Intell.

View full text Add to dashboard Cite

This paper introduces a novel approach to Item Response Theory (IRT) by incorporating deep learning to analyze student facial expressions to enhance the prediction and understanding of student responses to test items. This research is based on the assertion that students' facial expressions offer crucial insights into their cognitive and affective states during testing, subsequently influencing their item responses. The proposed State-Aware Deep Item Response Theory (SAD-IRT) model introduces a new parameter, the student state parameter, which can be viewed as a relative subjective difficulty parameter. It is latent-regressed from students' facial features while solving test items using state-of-the-art deep learning techniques. In an experiment with 20 students, SAD-IRT boosted prediction performance in students' responses compared to prior models without the student state parameter, including standard IRT and its deep neural network implementation, while maintaining consistent predictions of student ability and item difficulty parameters. The research further illustrates the model's early prediction ability in predicting the student's response result before the student answered. This study holds substantial implications for educational assessment, laying the groundwork for more personalized and effective learning and assessment strategies that consider students' emotional and cognitive states.

show abstract

Section: Methodsmentioning

confidence: 99%

State-Aware Deep Item Response Theory using student facial features

Zhou,

Suzuki,

Kumano

2024

Front. Artif. Intell.

View full text Add to dashboard Cite

show abstract

“…Empathetic characters that provide interventions generate superior results both to improve student interactions with the system, address negative student emotions, and in the overall learning experience [47]. Predicting outcomes of problems for students is a valuable source of information for planning and executing ITS interventions for improving learning [48], [49]. For example the ITS could provide hints when the system predicts that the student will not be able to successfully complete the problem.…”

Section: Related Workmentioning

confidence: 99%

“…Emotions such as frustration, boredom, and anxiety negatively influence learning outcomes of students [44]. To explore the correlation of emotion, engagement, and learning outcomes, we collected additional labels of student engagement on our MathSpringSP+ videos [48]. Specifically, we extract frames from videos in MathSpringSP+ dataset and annotate each frame with engagement labels (i.e., 'looking at their screen,' 'looking at their paper,' or 'wandering').…”

Section: Learning Outcome Based On Affect and Engagementmentioning

confidence: 99%

ATL-BP: A Student Engagement Dataset and Model for Affect Transfer Learning for Behavior Prediction

Ruiz

Allessio

et al. 2023

IEEE Trans. Biom. Behav. Identity Sci.

Self Cite

View full text Add to dashboard Cite

We propose a video-based transfer learning approach for predicting problem outcomes of students working with an intelligent tutoring system (ITS) by analyzing their faces and gestures. The ability to predict such outcomes enables tutoring systems to adjust interventions and ultimately yield improved student learning. We collected and released a labeled dataset of 2,749 problem-solving interaction samples of 54 students working with an intelligent online math tutor. Our transfer-learning challenge was then to design a representation in the source domain of images obtained from the Internet for facial expression analysis, and transfer this learned representation for human behavior prediction in the domain of webcam videos of students in a classroom environment. We developed a novel facial affect representation and a user-personalized training scheme that unlocks the potential of this representation. We designed several variants of a recurrent neural network that models the temporal structure of video sequences. Our final model, named ATL-BP for Affect Transfer Learning for Behavior Prediction, achieves a relative increase in the mean F-score of 50% over the state-of-the-art method on this new dataset. We also propose an additional set of annotations to predict students' engagement while solving a specific problem, and present models that can predict such engagement.

show abstract

“…A deep emotion classification model was trained through transfer learning from MobileNet [22]. The MobileNet pre-trained with ImageNet [23] was fine-tuned, and a similar architecture as Delgado [24] was used. The architecture added more layers from the original MobileNet, including one global average pooling 2D, one dense layer with 128 neurons with ReLU activation, and a 4-neuron classification with softmax activation.…”

Section: Deep Facial Emotion Recognition Model Setupmentioning

confidence: 99%

Exploring an Affective and Responsive Virtual Environment to Improve Remote Learning

Tang

Zhu

2023

Virtual Worlds

View full text Add to dashboard Cite

Online classes are typically conducted by using video conferencing software such as Zoom, Microsoft Teams, and Google Meet. Research has identified drawbacks of online learning, such as “Zoom fatigue”, characterized by distractions and lack of engagement. This study presents the CUNY Affective and Responsive Virtual Environment (CARVE) Hub, a novel virtual reality hub that uses a facial emotion classification model to generate emojis for affective and informal responsive interaction in a 3D virtual classroom setting. A web-based machine learning model is employed for facial emotion classification, enabling students to communicate four basic emotions live through automated web camera capture in a virtual classroom without activating their cameras. The experiment is conducted in undergraduate classes on both Zoom and CARVE, and the results of a survey indicate that students have a positive perception of interactions in the proposed virtual classroom compared with Zoom. Correlations between automated emojis and interactions are also observed. This study discusses potential explanations for the improved interactions, including a decrease in pressure on students when they are not showing faces. In addition, video panels in traditional remote classrooms may be useful for communication but not for interaction. Students favor features in virtual reality, such as spatial audio and the ability to move around, with collaboration being identified as the most helpful feature.

show abstract

Student Engagement Dataset

Cited by 16 publications

References 28 publications

State-Aware Deep Item Response Theory using student facial features

State-Aware Deep Item Response Theory using student facial features

ATL-BP: A Student Engagement Dataset and Model for Affect Transfer Learning for Behavior Prediction

Exploring an Affective and Responsive Virtual Environment to Improve Remote Learning

Contact Info

Product

Resources

About