Min-Te Sun
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Min-Te Sun has a total of 129 co-authors. Here are the top co-authors by number of shared publications:
- Wei-Shinn Ku
Auburn University
29 shared publications - Kazuya Sakai
Tokyo Metropolitan University
24 shared publications - Kotcharat Kitchat
National Central University
7 shared publications - Jie Wu
Cloud Computing Research Institute, China Telecom
5 shared publications - Quoc-Viet Nguyen
National Central University Department of Computer Science and Information Engineering
5 shared publications - Thattapon Surasak
King Mongkut’s University of Technology North Bangkok
5 shared publications
Latest Publications
- 20 October 2026
- Published by Emerald Publishing in Data Technologies and Applications
- p. 1-31
Smart questioning, answering, and clarifying to support adaptive learning
Purpose: Generative artificial intelligence (GAI) is increasingly integrated into higher education; however, many AI-based systems function primarily as reactive answer generators rather than adaptive instructional supports. This study develops and evaluates smart questioning, answering, and clarifying (SmartQAC), a pedagogically grounded AI-supported learning system that operationalizes structured dialogic scaffolding through questioning, answering, and clarification mechanisms to support adaptive learning. Design/methodology/approach: SmartQAC integrates retrieval-augmented generation, prompt-routing strategies, and structured dialogic interactions to promote metacognitive engagement and conceptual development. A quasi-experimental study involving 52 undergraduate students was conducted within a database course as an illustrative domain application. Mixed-method analyses were performed using pre-test and post-test assessments, interaction logs, questionnaires, and interviews to examine learning achievement, learning behaviors, and learner perceptions. Findings: Results indicate that students using SmartQAC achieved significantly higher learning achievement than those in the control group. Interaction behaviors within the SmartQAC environment were positively associated with students' learning achievement. Furthermore, participants reported positive perceptions regarding instructional usefulness, autonomy support, engagement, and pedagogical value. Originality/value: This study demonstrates how dialogic instructional principles can be operationalized within a retrieval-augmented large language model system to support adaptive learning. The findings provide empirical evidence for the effectiveness of integrating questioning, answering, and clarification mechanisms into AI-supported learning environments and offer practical design implications for future adaptive educational systems.
- 20 October 2026
- Published by Emerald Publishing in Data Technologies and Applications
- p. 1-31
Smart questioning, answering, and clarifying to support adaptive learning
Purpose: Generative artificial intelligence (GAI) is increasingly integrated into higher education; however, many AI-based systems function primarily as reactive answer generators rather than adaptive instructional supports. This study develops and evaluates smart questioning, answering, and clarifying (SmartQAC), a pedagogically grounded AI-supported learning system that operationalizes structured dialogic scaffolding through questioning, answering, and clarification mechanisms to support adaptive learning. Design/methodology/approach: SmartQAC integrates retrieval-augmented generation, prompt-routing strategies, and structured dialogic interactions to promote metacognitive engagement and conceptual development. A quasi-experimental study involving 52 undergraduate students was conducted within a database course as an illustrative domain application. Mixed-method analyses were performed using pre-test and post-test assessments, interaction logs, questionnaires, and interviews to examine learning achievement, learning behaviors, and learner perceptions. Findings: Results indicate that students using SmartQAC achieved significantly higher learning achievement than those in the control group. Interaction behaviors within the SmartQAC environment were positively associated with students' learning achievement. Furthermore, participants reported positive perceptions regarding instructional usefulness, autonomy support, engagement, and pedagogical value. Originality/value: This study demonstrates how dialogic instructional principles can be operationalized within a retrieval-augmented large language model system to support adaptive learning. The findings provide empirical evidence for the effectiveness of integrating questioning, answering, and clarification mechanisms into AI-supported learning environments and offer practical design implications for future adaptive educational systems.
- 1 May 2026
- Published by Springer Nature
- p. 242-259
UAV Path Planning for Building Structure Inspection
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Learning Technologies
- Vol. 19, 1034-1051
LLM-Based Multiagent System for Personalized Lesson Plan Generation in Multidomain Adaptive Learning
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Learning Technologies
- Vol. 19, 1034-1051
LLM-Based Multiagent System for Personalized Lesson Plan Generation in Multidomain Adaptive Learning
- 6 October 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 299
AI-enhanced structured dataset building for generating meaningful knowledge relationships and diversified questions
- 6 October 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 299
AI-enhanced structured dataset building for generating meaningful knowledge relationships and diversified questions
- 31 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 26 (11) , 18829-18838
Attention-Based Spatial-Temporal Graph Neural Network With Long-Term Dependencies for Traffic Speed Prediction
Urbanization, characterized by the continuous growth of population and density in urban areas, has led to the expansion and increased complexity of transportation networks, exacerbating traffic congestion. Accurate traffic speed prediction is crucial for effective traffic network management and planning. As the complexity of real-world road networks increases, integrating spatial and temporal information for accurate traffic speed prediction has become a challenging research task. This paper proposes a novel approach by introducing a spatial-temporal graph neural network (STGNN)-based model to enhance the accuracy of traffic speed prediction. By employing an attention-based STGNN, we effectively capture the complex relationships among road segments in real-world scenarios. We utilize the Huber loss as the training objective to improve prediction accuracy. Furthermore, we present an architecture that incorporates Root Mean Square Layer Normalization into the Transformer and integrates the Spatial-Temporal Attention Wavenet (STAWnet) into the model backbone, enabling faster training while maintaining model stability. We evaluate the proposed model using five real-world traffic speed benchmark datasets. The experimental results demonstrate that our method achieves superior performance compared to state-of-the-art traffic speed prediction approaches.
- 31 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 26 (11) , 18829-18838
Attention-Based Spatial-Temporal Graph Neural Network With Long-Term Dependencies for Traffic Speed Prediction
Urbanization, characterized by the continuous growth of population and density in urban areas, has led to the expansion and increased complexity of transportation networks, exacerbating traffic congestion. Accurate traffic speed prediction is crucial for effective traffic network management and planning. As the complexity of real-world road networks increases, integrating spatial and temporal information for accurate traffic speed prediction has become a challenging research task. This paper proposes a novel approach by introducing a spatial-temporal graph neural network (STGNN)-based model to enhance the accuracy of traffic speed prediction. By employing an attention-based STGNN, we effectively capture the complex relationships among road segments in real-world scenarios. We utilize the Huber loss as the training objective to improve prediction accuracy. Furthermore, we present an architecture that incorporates Root Mean Square Layer Normalization into the Transformer and integrates the Spatial-Temporal Attention Wavenet (STAWnet) into the model backbone, enabling faster training while maintaining model stability. We evaluate the proposed model using five real-world traffic speed benchmark datasets. The experimental results demonstrate that our method achieves superior performance compared to state-of-the-art traffic speed prediction approaches.
- 20 July 2025
- Published by Springer Nature
- p. 3-17
A Graph-Based System for PM2.5 Prediction
- 20 July 2025
- Published by Springer Nature
- p. 315-327
AI-Driven Classification of Chest Diseases Using X-Ray Images
- 20 July 2025
- Published by Springer Nature
- p. 315-327
AI-Driven Classification of Chest Diseases Using X-Ray Images
- 20 July 2025
- Published by Springer Nature
- p. 299-314
Advanced Golf Swing Analysis Using MediaPipe and Machine Learning
- 20 July 2025
- Published by Springer Nature
- p. 299-314
Advanced Golf Swing Analysis Using MediaPipe and Machine Learning
- 16 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cloud Computing
- Vol. 13 (2) , 721-736
Secure kNN for Distributed Cloud Environment Using Fully Homomorphic Encryption
Privacy-preserving k-nearest neighbor (PPkNN) classification for multiple clouds enables categorizing queried data into a class in keeping with data privacy, where the database and key servers jointly perform cryptographic operations. The existing solutions, unfortunately, take a long time and incur a large amount of traffic between the database and key servers. Therefore, in this paper, we propose a fast and secure kNN classification protocol, namely FSkNN, over distributed databases deployed in multiple clouds under the semi-honest model. Particularly, we focus on optimizing the network-related operations during kNN classification. That is, the proposed cryptographic protocol reduces the number of interactions between the servers by using a fully homomorphic encryption scheme and eliminates unnecessary traffic by applying mathematical techniques. In addition, the indistinguishability-based security of FSkNN is proven. We implemented FSkNN with C++ and the testbed experiments demonstrate that the proposed scheme significantly facilitates the query response time and reduces the communication cost.
- 16 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cloud Computing
- Vol. 13 (2) , 721-736
Secure kNN for Distributed Cloud Environment Using Fully Homomorphic Encryption
Privacy-preserving k-nearest neighbor (PPkNN) classification for multiple clouds enables categorizing queried data into a class in keeping with data privacy, where the database and key servers jointly perform cryptographic operations. The existing solutions, unfortunately, take a long time and incur a large amount of traffic between the database and key servers. Therefore, in this paper, we propose a fast and secure kNN classification protocol, namely FSkNN, over distributed databases deployed in multiple clouds under the semi-honest model. Particularly, we focus on optimizing the network-related operations during kNN classification. That is, the proposed cryptographic protocol reduces the number of interactions between the servers by using a fully homomorphic encryption scheme and eliminates unnecessary traffic by applying mathematical techniques. In addition, the indistinguishability-based security of FSkNN is proven. We implemented FSkNN with C++ and the testbed experiments demonstrate that the proposed scheme significantly facilitates the query response time and reduces the communication cost.
- 3 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Network Science and Engineering
- Vol. 12 (4) , 3206-3220
An Analysis of Anonymous MQTT for Publish-Subscribe-Based IoT Networks
Message queuing telemetry transport (MQTT) is widely used as a communication primitive in publish-subscribe-based IoT applications. However, the current MQTT standard does not support the privacy of IoT devices and users. Therefore, in this paper, we address the anonymous communications for MQTT-based IoT networks by tailored onion routing. To this end, we first introduce the concept of distributed IoT broker networks, in which a collection of IoT brokers are interconnected with each other. These brokers serve as onion routers. Then, we propose a generic anonymous messaging protocol (A-MQTT) for publish-subscribe-based IoT systems. Unlike typical anonymous communications, there are multiple destination IoT devices, called subscribers, in MQTT contexts. Hence, the privacy notions, such as traceable rate, device anonymity, and path anonymity, are redesigned. To illuminate the fundamental privacy issues, closed-form approximations for these privacy metrics are modeled. In addition, simulation results demonstrate that our A-MQTT significantly improves the security performance and that our proposed models closely approximate the various privacy performance. Furthermore, a proof-of-concept protocol is implemented on a Raspberry Pi 4 in order to show that the proposed A-MQTT can be deployed in an IoT environment.
- 3 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Network Science and Engineering
- Vol. 12 (4) , 3206-3220
An Analysis of Anonymous MQTT for Publish-Subscribe-Based IoT Networks
Message queuing telemetry transport (MQTT) is widely used as a communication primitive in publish-subscribe-based IoT applications. However, the current MQTT standard does not support the privacy of IoT devices and users. Therefore, in this paper, we address the anonymous communications for MQTT-based IoT networks by tailored onion routing. To this end, we first introduce the concept of distributed IoT broker networks, in which a collection of IoT brokers are interconnected with each other. These brokers serve as onion routers. Then, we propose a generic anonymous messaging protocol (A-MQTT) for publish-subscribe-based IoT systems. Unlike typical anonymous communications, there are multiple destination IoT devices, called subscribers, in MQTT contexts. Hence, the privacy notions, such as traceable rate, device anonymity, and path anonymity, are redesigned. To illuminate the fundamental privacy issues, closed-form approximations for these privacy metrics are modeled. In addition, simulation results demonstrate that our A-MQTT significantly improves the security performance and that our proposed models closely approximate the various privacy performance. Furthermore, a proof-of-concept protocol is implemented on a Raspberry Pi 4 in order to show that the proposed A-MQTT can be deployed in an IoT environment.