Development of an Unsupervised LearningBased Automated Evaluation System of Descriptive Assessment

• 발행기관: 한국교원대학교 뇌기반교육연구소
• 저자명: 하민수 최진선 임다니엘 최성철 주재걸 신철균 박지선
• 간행물 정보: 『Brain, Digital, & Learning』제13권 제4호, 339~351쪽, 전체 13쪽
• 주제분류: 사회과학 > 교육학
• 파일형태: PDF
• 발행일자: 2023.12.31

국문 초록

Current research on automatic scoring using traditional supervised learning methodscannot grade responses to questions that are newly generated or created spontaneously. Additionally, relying on pre-developed questions and their corresponding scoring modelsfor lesson planning may limit the creativity and diversity of instruction. This study proposesa method that can quickly evaluate student responses and generate feedback withoutthe need for pre-developed models. We introduces the SAAI system, which employsunsupervised learning techniques to instantly create scoring models based on studentresponses, thereby generating evaluation and feedback information. The SAAI systemcomplements the automatic scoring of traditional supervised learning methods andsupports scoring for a wide range of newly generated questions. This research elucidates theprinciples and significance of this system.

목차

Introduction

Overview of The Developed System

Method for System Development

Result and Discussion

Acknowledgements

Conflicts of Interest

References

키워드

unsupervised learning automatic evaluation descriptive assessment AI-based assessment