Artificial Intelligence in Education: 24th International Conference, AIED 2023, Tokyo, Japan, July 3–7, 2023, Proceedings

Preface
Organization
International Artificial Intelligence in Education Society
Contents
Full Papers
Machine-Generated Questions Attract Instructors When Acquainted with Learning Objectives
	1 Introduction
	2 Related Work
	3 Overview of Quadl
	4 Evaluation Study
		4.1 Model Implementation
		4.2 Survey Study
	5 Results
		5.1 Instructor Survey
		5.2 Accuracy of the Answer Prediction Model
		5.3 Qualitative Analysis of Questions Generated by Quadl
	6 Discussion
	7 Conclusion
	References
SmartPhone: Exploring Keyword Mnemonic with Auto-generated Verbal and Visual Cues
	1 Introduction
	2 Methodology
		2.1 Pipeline for Auto-generating Verbal and Visual Cues
	3 Experimental Evaluation
		3.1 Experimental Design
		3.2 Experimental Conditions
		3.3 Evaluation Metrics
		3.4 Results and Discussion
	4 Conclusions and Future Work
	References
Implementing and Evaluating ASSISTments Online Math Homework Support At large Scale over Two Years: Findings and Lessons Learned
	1 Introduction
	2 Background
		2.1 The ASSISTments Program
		2.2 Theoretical Framework
		2.3 Research Design
	3 Implementation of ASSISTments at Scale
		3.1 Recruitment
		3.2 Understanding School Context
		3.3 Training and Continuous Support
		3.4 Specifying a Use Model and Expectation
		3.5 Monitoring Dosage and Evaluating Quality of Implementation
	4 Data Collection
	5 Analysis and Results
	6 Conclusion
	References
The Development of Multivariable Causality Strategy: Instruction or Simulation First?
	1 Introduction
	2 Literature Review
		2.1 Learning Multivariable Causality Strategy with Interactive Simulation
		2.2 Problem Solving Prior to Instruction Approach to Learning
	3 Method
		3.1 Participants
		3.2 Design and Procedure
		3.3 Materials
		3.4 Data Sources and Analysis
	4 Results
	5 Discussion
	6 Conclusions, Limitations, and Future Work
	References
Content Matters: A Computational Investigation into the Effectiveness of Retrieval Practice and Worked Examples
	1 Introduction
	2 A Computational Model of Human Learning
	3 Simulation Studies
		3.1 Data
		3.2 Method
	4 Results
		4.1 Pretest
		4.2 Learning Gain
		4.3 Error Type
	5 General Discussion
	6 Future Work
	7 Conclusions
	References
Investigating the Utility of Self-explanation Through Translation Activities with a Code-Tracing Tutor
	1 Introduction
		1.1 Code Tracing: Related Work
	2 Current Study
		2.1 Translation Tutor vs. Standard Tutor
		2.2 Participants
		2.3 Materials
		2.4 Experimental Design and Procedure
	3 Results
	4 Discussion and Future Work
	References
Reducing the Cost: Cross-Prompt Pre-finetuning for Short Answer Scoring
	1 Introduction
	2 Related Work
	3 Preliminaries
		3.1 Task Definition
		3.2 Scoring Model
	4 Method
	5 Experiment
		5.1 Dataset
		5.2 Setting
		5.3 Results
		5.4 Analysis: What Does the SAS Model Learn from Pre-finetuning on Cross Prompt Data?
	6 Conclusion
	References
Go with the Flow: Personalized Task Sequencing Improves Online Language Learning
	1 Introduction
	2 Related Work
		2.1 Adaptive Item Sequencing
		2.2 Individual Adjustment of Difficulty Levels in Language Learning
	3 Methodology
		3.1 Online-Controlled Experiment
	4 Results
		4.1 H1 – Incorrect Answers
		4.2 H2 – Dropout
		4.3 H3 – User Competency
	5 Discussion
	6 Conclusion
	References
Automated Hand-Raising Detection in Classroom Videos: A View-Invariant and Occlusion-Robust Machine Learning Approach
	1 Introduction
	2 Related Work
	3 Methodology
		3.1 Data
		3.2 Skeleton-Based Hand-Raising Detection
		3.3 Automated Hand-Raising Annotation
	4 Results
		4.1 Relation Between Hand-Raising and Self-reported Learning Activities
		4.2 Hand-Raising Classification
		4.3 Automated Hand-Raising Annotation
	5 Discussion
	6 Conclusion
	References
Robust Educational Dialogue Act Classifiers with Low-Resource and Imbalanced Datasets
	1 Introduction
	2 Background
		2.1 Educational Dialogue Act Classification
		2.2 AUC Maximization on Imbalanced Data Distribution
	3 Methods
		3.1 Dataset
		3.2 Scheme for Educational Dialogue Act
		3.3 Approaches for Model Optimization
		3.4 Model Architecture by AUC Maximization
		3.5 Study Setup
	4 Results
		4.1 AUC Maximization Under Low-Resource Scenario
		4.2 AUC Maximization Under Imbalanced Scenario
	5 Discussion and Conclusion
	References
What and How You Explain Matters: Inquisitive Teachable Agent Scaffolds Knowledge-Building for Tutor Learning
	1 Introduction
	2 SimStudent: The Teachable Agent
	3 Constructive Tutee Inquiry
		3.1 Motivation
		3.2 Response Classifier
		3.3 Dialog Manager
	4 Method
	5 Results
		5.1 RQ1: Can we Identify Knowledge-Building and Knowledge-Telling from Tutor Responses to Drive CTI?
		5.2 RQ2: Does CTI Facilitate Tutor Learning?
		5.3 RQ3: Does CTI Help Tutors Learn to Engage in Knowledge-Building?
	6 Discussion
	7 Conclusion
	References
Help Seekers vs. Help Accepters: Understanding Student Engagement with a Mentor Agent
	1 Introduction
	2 Mr. Davis and Betty’s Brain
	3 Methods
		3.1 Participants
		3.2 Interaction Log Data
		3.3 In-situ Interviews
		3.4 Learning and Anxiety Measures
	4 Results
		4.1 Help Acceptance
		4.2 Help Seeking
		4.3 Learning Outcomes
		4.4 Insights from Qualitative Interviews
	5 Conclusions
	References
Adoption of Artificial Intelligence in Schools: Unveiling Factors Influencing Teachers’ Engagement
	1 Introduction
	2 Context and the Adaptive Learning Platform Studied
	3 Methodology
	4 Results
		4.1 Teachers’ Responses to the Items
		4.2 Predicting Teachers’ Engagement with the Adaptive Learning Platform
	5 Discussion
	6 Conclusion
	Appendix
	References
The Road Not Taken: Preempting Dropout in MOOCs
	1 Introduction
	2 Related Work
	3 Method
		3.1 Dataset
		3.2 Modeling Student Engagement by HMM
		3.3 Study Setup
	4 Results
	5 Discussion and Conclusion
	References
Does Informativeness Matter? Active Learning for Educational Dialogue Act Classification
	1 Introduction
	2 Related Work
		2.1 Educational Dialogue Act Classification
		2.2 Sample Informativeness
		2.3 Statistical Active Learning
	3 Methods
		3.1 Dataset
		3.2 Educational Dialogue Act Scheme and Annotation
		3.3 Identifying Sample Informativeness via Data Maps
		3.4 Active Learning Selection Strategies
		3.5 Study Setup
	4 Results
		4.1 Estimation of Sample Informativeness
		4.2 Efficacy of Statistical Active Learning Methods
	5 Conclusion
	References
Can Virtual Agents Scale Up Mentoring?: Insights from College Students' Experiences Using the CareerFair.ai Platform at an American Hispanic-Serving Institution
	1 Introduction
	2 CareerFair.ai Design
	3 Research Design
	4 Results
	5 Discussion
	6 Conclusions and Future Directions
	References
Real-Time AI-Driven Assessment and Scaffolding that Improves Students’ Mathematical Modeling during Science Investigations
	1 Introduction
		1.1 Related Work
	2 Methods
		2.1 Participants and Materials
		2.2 Inq-ITS Virtual Lab Activities with Mathematical Modeling
		2.3 Approach to Automated Assessment and Scaffolding of Science Practices
		2.4 Measures and Analyses
	3 Results
	4 Discussion
	References
Improving Automated Evaluation of Student Text Responses Using GPT-3.5 for Text Data Augmentation
	1 Introduction
	2 Background and Research Questions
	3 Methods
		3.1 Data Sets
		3.2 Augmentation Approach
		3.3 Model Classification
		3.4 Baseline Evaluation
	4 Results
	5 Discussion
	6 Conclusion
	7 Future Work
	References
The Automated Model of Comprehension Version 3.0: Paying Attention to Context
	1 Introduction
	2 Method
		2.1 Processing Flow
		2.2 Features Derived from AMoC
		2.3 Experimental Setup
		2.4 Comparison Between AMoC Versions
	3 Results
		3.1 Use Case
		3.2 Correlations to the Landscape Model
		3.3 Diffentiating Between High-Low Cohesion Texts
	4 Conclusions and Future Work
	References
Analysing Verbal Communication in Embodied Team Learning Using Multimodal Data and Ordered Network Analysis
	1 Introduction
	2 Methods
	3 Results
		3.1 Primary Tasks
		3.2 Secondary Tasks
	4 Discussion
	References
Improving Adaptive Learning Models Using Prosodic Speech Features
	1 Introduction
	2 Methods
		2.1 Participants
		2.2 Design and Procedure
		2.3 Materials
		2.4 Speech Feature Extraction
		2.5 Data and Statistical Analyses
	3 Results
		3.1 The Association Between Speech Prosody and Memory Retrieval Performance
		3.2 Improving Predictions of Future Performance Using Speech Prosody
	4 Discussion
	References
Neural Automated Essay Scoring Considering Logical Structure
	1 Introduction
	2 Conventional Neural AES Model Using BERT
	3 Argument Mining
	4 Proposed Method
		4.1 DNN Model for Processing Logical Structure
		4.2 Neural AES Model Considering Logical Structure
	5 Experiment
		5.1 Setup
		5.2 Experimental Results
		5.3 Analysis
	6 Conclusion
	References
“Why My Essay Received a 4?”: A Natural Language Processing Based Argumentative Essay Structure Analysis
	1 Introduction
	2 Literature Review
		2.1 Automatic Essay Scoring
		2.2 Argument Mining
	3 Data
		3.1 Feedback Prize Dataset
		3.2 ACT Writing Test Dataset
	4 System Design
		4.1 Datasets
		4.2 Ensemble Model Block
		4.3 Essay Analysis Block
	5 Results
		5.1 Ensemble Model Results
		5.2 ACT Tests Essays Analysis Results
		5.3 The Feedback Proving Process
	6 Discussion
	7 Conclusion
	Appendix
	References
Leveraging Deep Reinforcement Learning for Metacognitive Interventions Across Intelligent Tutoring Systems
	1 Introduction
	2 Background and Related Work
		2.1 Metacognitive Interventions for Strategy Instruction
		2.2 Reinforcement Learning in Intelligent Tutoring Systems
	3 Logic and Probability Tutors
	4 Methods
		4.1 Experiment 1: RFC-Static
		4.2 Experiment 2: DRL-Adaptive
	5 Experiments Setup
	6 Results
		6.1 Experiment 1: RFC-Static
		6.2 Experiment 2: DRL-Adaptive
		6.3 Post-hoc Analysis
	7 Discussions and Conclusions
	References
Enhancing Stealth Assessment in Collaborative Game-Based Learning with Multi-task Learning
	1 Introduction
	2 Related Work
	3 Dataset
		3.1 Out-of-Domain Labeling
		3.2 Post-test Assessment
		3.3 Feature Extraction
		3.4 Class Labeling
	4 Model Architecture
	5 Results
	6 Discussion
	7 Conclusion
	References
How Peers Communicate Without Words-An Exploratory Study of Hand Movements in Collaborative Learning Using Computer-Vision-Based Body Recognition Techniques
	1 Introduction
	2 Literature Review
	3 Method
		3.1 Participants and Learning Context
		3.2 Data Collection
		3.3 Data Analysis and Instruments
	4 Results
	5 Discussion
	References
Scalable Educational Question Generation with Pre-trained Language Models
	1 Introduction
	2 Related Work
		2.1 Automatic Question Generation (QG)
		2.2 Pre-trained Language Models (PLMs) for Educational QG
		2.3 Related Datasets
	3 Methodology
		3.1 Research Questions
		3.2 Question Generations Models
		3.3 Data
		3.4 Evaluation Metrics
		3.5 Experimental Setup
	4 Results
	5 Discussion
		5.1 Ability of PLMs to Generate Educational Questions (RQ1)
		5.2 Effect of Pre-training with a Scientific Text Corpus (RQ2)
		5.3 Impact of the Training Size on the Question Quality (RQ3)
		5.4 Effect of Fine-Tuning Using Educational Questions (RQ4)
		5.5 Opportunities
		5.6 Limitations
	6 Conclusion
	References
Involving Teachers in the Data-Driven Improvement of Intelligent Tutors: A Prototyping Study
	1 Introduction
	2 Needs-Finding Study
	3 SolutionVis
	4 User Study
	5 Results
	6 Discussion
	7 Conclusion
	References
Reflexive Expressions: Towards the Analysis of Reflexive Capability from Reflective Text
	1 Introduction
	2 From Reflective Properties, to Reflexive Interactions
	3 Reflexivity and Reflective Writing Analytics
	4 Methodology
		4.1 Phase T - Theoretical Categories
		4.2 Phase C - Computational Ngrams
		4.3 Phase V - Verification Judgements
	5 Findings and Discussion
	6 Limitations and Future Work
	7 Conclusion
	References
Algebra Error Classification with Large Language Models
	1 Introduction
	2 Methodology
		2.1 The Algebra Error Classification Task
		2.2 Our Method
	3 Experimental Evaluation
		3.1 Dataset Details
		3.2 Metrics and Baselines
		3.3 Implementation Details
		3.4 Results and Analysis
	4 Discussion, Conclusion, and Future Work
	References
Exploration of Annotation Strategies for Automatic Short Answer Grading
	1 Introduction
	2 Related Work
	3 Entailment Based Answer Grading
		3.1 Problem Formulation
		3.2 Model Description
		3.3 Fine-Tuning the ASAG Model
	4 Annotation Strategies
	5 Experimental Setting
	6 Few-Shot Experiments
	7 Cross-Domain Experiments
	8 Comparison to the State-of-the-Art
	9 Conclusion
	References
Impact of Learning a Subgoal-Directed Problem-Solving Strategy Within an Intelligent Logic Tutor
	1 Introduction
	2 Related Work
	3 Method
		3.1 Deep Thought (DT), the Intelligent Logic Tutor
		3.2 Experiment Design
	4 RQ1 (Students' Experience): Difficulties Across MPS and PS Problems
	5 RQ2: Students' Performance After Training
	6 RQ3: Proof-Construction and Subgoaling Approach/Skills After Training
		6.1 Student Approaches in Training-Level Test Problems
		6.2 Student Approaches in Posttest Problems
	7 Discussion
	8 Conclusion and Future Work
	References
Matching Exemplar as Next Sentence Prediction (MeNSP): Zero-Shot Prompt Learning for Automatic Scoring in Science Education
	1 Introduction
	2 Related Work
		2.1 Natural Language Processing for Automatic Scoring
		2.2 Prompt Learning
	3 Approach
		3.1 Matching Exemplars
		3.2 Zero Grade Identifier
	4 Experiment
		4.1 Setup
		4.2 Results
	5 Conclusion and Discussion
	References
Learning When to Defer to Humans for Short Answer Grading
	1 Introduction
	2 Related Work
	3 Description of the Data
	4 Methods
	5 Results
	6 Discussion and Conclusion
	References
Contrastive Learning for Reading Behavior Embedding in E-book System
	1 Introduction
	2 Related Work
	3 Contrastive Learning for Reading Behavior Embedding
		3.1 Reading Log Segmentation
		3.2 Absolute and Relative Time-Positional Encoding
		3.3 Network Architecture
		3.4 Network Training
	4 Experimental Settings
		4.1 At-Risk Student Detection Settings
		4.2 Dataset and Parameter Settings
	5 Experimental Results
		5.1 Evaluation of At-Risk Student Detection
		5.2 CRE Feature Space Analysis
	6 Conclusion and Discussion
	References
Generalizable Automatic Short Answer Scoring via Prototypical Neural Network
	1 Introduction
	2 Related Work
	3 Method
	4 Experiment
		4.1 Data
		4.2 Baseline Methods
		4.3 Training, Evaluation, Metrics and Implementation Details
	5 Results
	6 Conclusion and Future Work
	References
A Spatiotemporal Analysis of Teacher Practices in Supporting Student Learning and Engagement in an AI-Enabled Classroom
	1 Introduction
		1.1 Spatiotemporal Factors in Teacher Practices with AI Tutors
		1.2 Research Questions and Hypotheses
	2 Methods
		2.1 Case Study Context
		2.2 Teacher Visits in the Temporal Context of Student Learning
	3 Results
		3.1 RQ1: Factors Associated with Teacher’s Choice of Students to Visit
		3.2 RQ2: Teacher Visit Associations with Student Engagement and Learning
	4 Discussion and Conclusion
	References
Trustworthy Academic Risk Prediction with Explainable Boosting Machines
	1 Introduction
	2 Background
		2.1 Academic Risk Prediction
		2.2 Explainable AI in Education
		2.3 Explainable Boosting Machines
	3 Methods
		3.1 Study Area and Data
		3.2 Training
		3.3 Model Assessment
	4 Experimental Results
		4.1 Explainability of EBMs
		4.2 Accuracy
		4.3 Earliness and Stability
		4.4 Fairness
		4.5 Faithfulness of Explanations
		4.6 Discussion
	5 Conclusion and Outlook
	References
Automatic Educational Question Generation with Difficulty Level Controls
	1 Introduction
	2 Related Work
	3 Problem Formulation
	4 Approach
		4.1 Age of Acquisition Based Sampling
		4.2 Energy Components
	5 Experiments
		5.1 Data Preparation
		5.2 Experiment Settings
		5.3 Expert Models
		5.4 Baselines
		5.5 Question Quality Evaluations and Observations
		5.6 Difficulty Controllability Analysis
	6 User Study
	7 Conclusion
	References
Getting the Wiggles Out: Movement Between Tasks Predicts Future Mind Wandering During Learning Activities
	1 Introduction
		1.1 Background and Related Work
		1.2 Current Work: Contributions and Novelty
	2 Methods
		2.1 Data Collection
		2.2 Machine Learning Models
		2.3 Sensors, Data Processing, and Feature Extraction
	3 Results
		3.1 Model Comparisons
		3.2 Predictive Features
	4 Discussion
	References
Efficient Feedback and Partial Credit Grading for Proof Blocks Problems
	1 Introduction
	2 Related Work
		2.1 Software for Learning Mathematical Proofs
		2.2 Edit Distance Based Grading and Feedback
	3 Proof Blocks
	4 The Edit Distance Algorithm
		4.1 Mathematical Preliminaries
		4.2 Problem Definition
		4.3 Baseline Algorithm
		4.4 Optimized (MVC-Based) Implementation of Edit Distance Algorithm
		4.5 Worked Example of Algorithm 2
		4.6 Proving the Correctness of Algorithm 2
	5 Benchmarking Algorithms on Student Data
		5.1 Data Collection
		5.2 Benchmarking Details
		5.3 Results
	6 Conclusions and Future Work
	References
Dropout Prediction in a Web Environment Based on Universal Design for Learning
	1 Introduction
	2 Related Work
	3 Research Questions
	4 Methodology
		4.1 The Learning Platform I3Learn and Dropout Level
		4.2 Data Collection and Features
		4.3 Data Aggregation
	5 Dataset
	6 Results
		6.1 RQ 1: Transfer of Methods for Dropout Prediction
		6.2 RQ 2: Dropout Prediction with Data of Diverse Granularity
		6.3 RQ 3: Assessments for Predicting Dropout
	7 Conclusion
	References
Designing for Student Understanding of Learning Analytics Algorithms
	1 Introduction
	2 Prior Work
	3 Knowledge Components of BKT
	4 The BKT Interactive Explanation
	5 Impact of Algorithmic Transparency on User Understanding and Perceptions
	6 Conclusion
	References
Feedback and Open Learner Models in Popular Commercial VR Games: A Systematic Review
	1 Introduction
	2 Background
	3 Methods: Game Selection and Coding
	4 Results
	5 Discussion and Conclusions
	References
Gender Differences in Learning Game Preferences: Results Using a Multi-dimensional Gender Framework
	1 Introduction
	2 Methods
		2.1 Participants
		2.2 Materials and Procedures
	3 Results
		3.1 Game Genre Preferences
		3.2 Game Narrative Preferences
		3.3 Post-hoc Analyses
	4 Discussion and Conclusion
	References
Can You Solve This on the First Try? – Understanding Exercise Field Performance in an Intelligent Tutoring System
	1 Introduction
	2 Related Work
		2.1 Academic Performance Prediction
		2.2 Investigating Factors that Influence Academic Performance
		2.3 Explainable AI for Academic Performance Prediction
	3 Methodology
		3.1 Dataset
		3.2 Data Analysis
	4 Results
	5 Discussion and Conclusion
	References
A Multi-theoretic Analysis of Collaborative Discourse: A Step Towards AI-Facilitated Student Collaborations
	1 Introduction
		1.1 Background and Related Work
	2 Methods
	3 Results and Discussion
	4 General Discussion
	References
Development and Experiment of Classroom Engagement Evaluation Mechanism During Real-Time Online Courses
	1 Introduction
	2 Related Work
		2.1 Behavior Estimation
		2.2 Class Evaluation System
	3 Online Education Platform
	4 Multi-reaction Estimation
		4.1 Student Head Reaction
		4.2 Student Expression Reaction
	5 Online Classroom Evaluation
	6 Experiments
		6.1 Experiment I: Instruction Experiment
		6.2 Experiment II: Simulation Classroom Experiment
	7 Discussion
	8 Conclusion
	References
Physiological Synchrony and Arousal as Indicators of Stress and Learning Performance in Embodied Collaborative Learning
	1 Introduction
	2 Background and Related Work
	3 Methods
		3.1 Educational Context
		3.2 Apparatus and Data Collection
		3.3 Feature Engineering
		3.4 Analysis
	4 Results
	5 Discussion
	6 Conclusion
	References
Confusion, Conflict, Consensus: Modeling Dialogue Processes During Collaborative Learning with Hidden Markov Models
	1 Introduction and Related Work
	2 Methods
	3 Results
	4 Discussion
		4.1 Dialogue States
		4.2 Transitions into and Out of Exploratory Talk
		4.3 Design Implications
	5 Conclusion
	References
Unsupervised Concept Tagging of Mathematical Questions from Student Explanations
	1 Introduction
	2 Related Work
	3 Experiment Setup
	4 Unsupervised Question Tagging Based on Student Explanations
		4.1 Manual Tagging Based on Drag-and-Drop Activity
		4.2 Our Method: Unsupervised Skill Tagging (UST)
	5 Results and Discussion
	6 Conclusion and Future Work
	References
Robust Team Communication Analytics with Transformer-Based Dialogue Modeling
	1 Introduction
	2 Related Work
	3 Team Communication
	4 Dataset
	5 Team Communication Analysis Framework
	6 Evaluation
	7 Conclusion
	References
Teacher Talk Moves in K12 Mathematics Lessons: Automatic Identification, Prediction Explanation, and Characteristic Exploration
	1 Introduction
	2 Related Work
		2.1 Automated Models on Classroom Discourse
		2.2 Explainable Artificial Intelligence
	3 Method
		3.1 Dataset
		3.2 Model Construction
		3.3 Model Explanation
	4 Experiments and Results
		4.1 Interpreting Results Validation
		4.2 Talk Move Characteristics Exploration
	5 Discussion and Conclusion
	References
Short Papers
Ethical and Pedagogical Impacts of AI in Education
	1 Introduction
	2 Research Method
	3 Results, Discussion and Practical Implication
		3.1 Learner Status
		3.2 Learning Environment and Experience
		3.3 Educational Processes and Approaches
		3.4 Interaction, Pedagogical Relationship, and Roles
	4 Conclusion
	References
Multi-dimensional Learner Profiling by Modeling Irregular Multivariate Time Series with Self-supervised Deep Learning
	1 Introduction
	2 Approach
	3 Experiments
	4 Conclusion
	References
Examining the Learning Benefits of Different Types of Prompted Self-explanation in a Decimal Learning Game
	1 Introduction
	2 A Digital Learning Game for Decimal Numbers
	3 Methods
	4 Results
	5 Discussion and Conclusion
	References
Plug-and-Play EEG-Based Student Confusion Classification in Massive Online Open Courses
	1 Introduction
	2 Dataset
	3 Methodology
	4 Results
	5 Conclusion
	References
CPSCoach: The Design and Implementation of Intelligent Collaborative Problem Solving Feedback
	1 Introduction and Related Work
	2 Intervention Design and User Study
	3 Results and Discussion
	References
Development of Virtual Reality SBIRT Skill Training with Conversational AI in Nursing Education
	1 Introduction
		1.1 Conversational AI in Healthcare Education
	2 Interaction Design and Implementation of the Application
		2.1 SBIRT Conversation Data Collection
		2.2 Design of Interaction Modes
	3 Mode 3: Virtual Patient Conversational AI Mode.
	4 Conclusion
	References
Balancing Test Accuracy and Security in Computerized Adaptive Testing
	1 Introduction
	2 Methodology
		2.1 BOBCAT Background
		2.2 C-BOBCAT
	3 Experiments
		3.1 Data, Experimental Setup, Baseline, and Evaluation Metrics
		3.2 Results and Discussion
	4 Conclusions and Future Work
	References
A Personalized Learning Path Recommendation Method for Learning Objects with Diverse Coverage Levels
	1 Introduction
	2 The Learning Path Recommendation Framework
	3 The Graph-Based Genetic Algorithm
		3.1 Problem Modeling
		3.2 Chromosome Representation
		3.3 The Genetic Operators
	4 Experiments
	5 Conclusion
	References
Prompt-Independent Automated Scoring of L2 Oral Fluency by Capturing Prompt Effects
	1 Introduction
	2 Related Work
	3 Prompt-Independent Automated Fluency Scoring
	4 Experiment
	5 Results and Discussion
	References
Navigating Wanderland: Highlighting Off-Task Discussions in Classrooms
	1 Introduction
	2 Methodology
		2.1 Dataset
		2.2 Models
	3 Results
	4 Conclusion
	References
C2Tutor: Helping People Learn to Avoid Present Bias During Decision Making
	1 Introduction
	2 Literature Review and Research Gaps
	3 C2Tutor - Reducing Present Bias
	4 Experimental Design
	5 Findings
	6 Conclusion and Future Work
	References
A Machine-Learning Approach to Recognizing Teaching Beliefs in Narrative Stories of Outstanding Professors
	1 Introduction
	2 Related Work
	3 Dataset
	4 Methodology
	5 Evaluation
		5.1 Evaluation Setting
		5.2 Comparative Method
		5.3 Evaluation Result
	6 Conclusion
	References
BETTER: An Automatic feedBack systEm for supporTing emoTional spEech tRaining
	1 Introduction
	2 Related Work
	3 Methodology
		3.1 BETTER Version 1.0
		3.2 Preliminary Experiment
		3.3 BETTER Version 2.0
	4 Conclusion
	References
Eliciting Proactive and Reactive Control During Use of an Interactive Learning Environment
	1 Introduction
	2 Task Design
	3 Study 1: Inducing Proactive and Reactive Control
	4 Study 2: Shifting from Proactive to Reactive Control
	5 Discussion
	References
How to Repeat Hints: Improving AI-Driven Help in Open-Ended Learning Environments
	1 Introduction
	2 UnityCT
	3 User Study
	4 Analysis and Results
	5 Discussion and Future Work
	References
Automatic Detection of Collaborative States in Small Groups Using Multimodal Features
	1 Introduction
	2 Methods
		2.1 Data Collection
		2.2 Annotations
		2.3 Verbal Features
		2.4 Prosodic Features
		2.5 Model Training
	3 Results
	4 Discussion
		4.1 Qualitative Error Analysis
	5 Limitations, Future Work, and Conclusion
	References
Affective Dynamic Based Technique for Facial Emotion Recognition (FER) to Support Intelligent Tutors in Education
	1 Introduction
	2 Proposed Method
		2.1 Affective Dynamics Model
		2.2 Affective Dynamics Based FER Technique
	3 Experimental Evaluations
		3.1 Results
	4 Conclusion
	References
Real-Time Hybrid Language Model for Virtual Patient Conversations
	1 Introduction
	2 Related Works
	3 Dataset
	4 Methodology
	5 Results and Discussion
	6 Conclusion
	References
Towards Enriched Controllability for Educational Question Generation
	1 Introduction
	2 Generating Explicit and Implicit Questions
	3 Experimental Setup
	4 Results
	5 Conclusion
	References
A Computational Model for the ICAP Framework: Exploring Agent-Based Modeling as an AIED Methodology
	1 Introduction
	2 ABICAP
		2.1 Learning
	3 Simulations and Results
	4 Discussion
	References
Automated Program Repair Using Generative Models for Code Infilling
	1 Introduction
	2 Background
	3 Methodology
	4 Results
	5 Discussion and Conclusion
	References
Measuring the Quality of Domain Models Extracted from Textbooks with Learning Curves Analysis
	1 Introduction
	2 Background
	3 Experiment
	4 Results and Analysis
	5 Conclusion and Future Work
	References
Predicting Progress in a Large-Scale Online Programming Course
	1 Introduction
	2 Background and Related Work
	3 Data
	4 Method
		4.1 Slide Interaction Data Extraction and Train-Test Split
		4.2 Feature Selection and Ranking
		4.3 Classification
	5 Results
		5.1 Evaluation by Educators
	6 Discussion
		6.1 Accuracy in Predicting End of Module Outcomes
		6.2 Effects of Feature Selection
		6.3 Key Observations on Chosen Slides
	7 Conclusion
	References
Examining the Impact of Flipped Learning for Developing Young Job Seekers’ AI Literacy
	1 Introduction
	2 Methodology
		2.1 Participants and Research Design
		2.2 Data Collection and Analysis
	3 Results
		3.1 Academic Achievement
		3.2 Educational Satisfaction
		3.3 Focus Group Interview
	4 Discussion and Conclusion
	References
Automatic Analysis of Student Drawings in Chemistry Classes
	1 Introduction
	2 Automatic Categorization of Student Drawings
		2.1 Automatic Dataset Creation
		2.2 Detection of Objects in Student Drawings
		2.3 Classification of Drawing Characteristics
	3 Experimental Evaluation
		3.1 Datasets
		3.2 Detection of Objects
		3.3 Classification of Drawing Characteristics Used by Students
	4 Conclusions
	References
Training Language Models for Programming Feedback Using Automated Repair Tools
	1 Introduction
	2 Related Work
	3 Methodology
	4 Results and Discussion
	5 Conclusion
	References
Author Index