แจ้งเอกสารไม่ครบถ้วน, ไม่ตรงกับชื่อเรื่อง หรือมีข้อผิดพลาดเกี่ยวกับเอกสาร ติดต่อที่นี่ ==>
หากไม่มีอีเมลผู้รับให้กรอก thailis-noc@uni.net.th
หากไม่มีอีเมลผู้รับให้กรอก thailis-noc@uni.net.th
Peerasak Pianprasit. A framework for prompt-based learning automated feedback system. (). King Mongkut's University of Technology North Bangkok. Central Library. : , 2026.
A framework for prompt-based learning automated feedback system
Name: Wittawas Puntumchinda
Organization :
Burapha University. Faculty of Informatics
Email :
wittawas@buu.ac.th
Name: Waranrach Viriyavit
Organization :
Burapha University. Faculty of Informatics
Email :
waranrach.vi@go.buu.ac.th
Name: Suwanna Rasmequan
Organization :
Burapha University. Faculty of Informatics
Email :
rsuwanna@buu.ac.th
Name: Peerasak Pianprasit
Organization :
Burapha University. Faculty of Informatics
Email :
peerasak@buu.ac.th
keyword:
Large language models
LCSH:
Feedback (Psychology)
-- Data processing
Abstract:
Large language models (LLMs) have enabled automated feedback systems capable of generating context-aware and pedagogically meaningful responses. However, ensuring the reliability and consistency of AI-generated feedback remains a significant challenge. This paper presents a human-in-the-loop prompt-based framework for improving the semantic alignment between LLM-generated feedback and human instructor evaluation. The framework integrates instructor-defined prompting strategies and evaluates their effectiveness using semantic similarity as a proxy for pedagogical fidelity. An experimental workflow was designed in which a human instructor first evaluated student programming assignments to establish reference feedback. The instructor then created zero-shot, one-shot, and few-shot prompts respectively to guide an Ollama-based LLM in generating automated feedback. The semantic similarity between AI-generated and instructor-provided feedback was computed across eight programming tasks representing varying levels of complexity and code quality. Results show that few-shot prompting consistently improves alignment with human feedback, achieving an average performance gain of 3.6 percent (SD = 3.01) compared to zero-shot and one-shot approaches. Performance improvements were observed in six of eight tasks, with one task showing no change and one exhibiting a slight decline, indicating task dependency. The results indicate that integrating structured exemplar-based prompting within a human-in-the-loop framework significantly improves semantic alignment, reliability, and pedagogical consistency in LLM-based automated feedback systems, underscoring the critical role of prompt engineering and the effectiveness of structured in-context examples in producing consistent, trustworthy, and instructionally aligned feedback.
King Mongkut's University of Technology North Bangkok. Central Library
Address:
BANGKOK
Email:
library@kmutnb.ac.th
Created:
2026
Modified:
2026-08-11
Issued:
2026-08-11
บทความ/Article
application/pdf
BibliograpyCitation :
In ECTI Association and Burapha University. Faculty of Engineering. 2026 23rd International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON 2026) (pp.416-421). Piscataway, NJ : IEEE
eng
©copyrights King Mongkut's University of Technology North Bangkok
RightsAccess:
ใช้เวลา
0.038936 วินาที
Wittawas Puntumchinda
| Title | Contributor | Type |
|---|---|---|
| A framework for prompt-based learning automated feedback system
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Wittawas Puntumchinda;Waranrach Viriyavit;Suwanna Rasmequan;Peerasak Pianprasit | บทความ/Article |
Waranrach Viriyavit
| Title | Contributor | Type |
|---|---|---|
| A framework for prompt-based learning automated feedback system
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Wittawas Puntumchinda;Waranrach Viriyavit;Suwanna Rasmequan;Peerasak Pianprasit | บทความ/Article | |
| Synthetic Thai government dataset for privacy-compliant ocr vision llms development
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Waranrach Viriyavit;Yanakorn Bunphanuk;Chaianun Damrongrat;Sarawoot Kongyoung | บทความ/Article |
Suwanna Rasmequan
| Title | Contributor | Type |
|---|---|---|
| A framework for prompt-based learning automated feedback system
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Wittawas Puntumchinda;Waranrach Viriyavit;Suwanna Rasmequan;Peerasak Pianprasit | บทความ/Article |
Peerasak Pianprasit
| Title | Contributor | Type |
|---|---|---|
| KruCode: an online learning programming platform for programming course
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Nuttaporn Phakdee;Peerasak Pianprasit | บทความ/Article | |
| Breast tumor classification in ultrasound images using gaussian noise-based feature augmentation
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Ungsumalee Suttapakti;Wich Chanchalermchai;Peerasak Pianprasit;Kanuengnij Kubola;Pichet Wayalun | บทความ/Article | |
| A framework for prompt-based learning automated feedback system
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Wittawas Puntumchinda;Waranrach Viriyavit;Suwanna Rasmequan;Peerasak Pianprasit | บทความ/Article |