초등 저학년 쓰기 학습 부진 판별 인공지능 모델 탐색
Exploring AI Models to Discriminate Writing Difficulties in the Lower Elementary School Children
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
본 연구의 목적은 초등 저학년 학생의 쓰기 학습 부진의 조기 진단 및 예측을 위해 손글씨 이미지 분 류를 통한 판별 인공지능 모델을 개발하여 정확도를 탐색하는 것이다. 초등학교 1학년생들의 쓰기 수행 과정 관찰을 통한 성취 수준을 쓰기 보통, 쓰기 부진으로 구분한 후, 쌍자음과 겹받침, 이중모음을 고려 해 선정된 어휘의 손글씨 이미지 자료를 수집하였다. 수집된 손글씨 데이터에 대하여 어휘 특징별 두 학 습 집단으로 구분하여, 엔트리 AI 블록 기반의 쓰기 학습 부진 판별 인공지능 모델을 개발하였다. 개발 된 인공지능 모델별 판별 결과의 정확도를 비교한 결과, 받침이 많은 어휘로 학습한 모델이 가장 정확도 가 높았다. 이를 통해 초등학교 저학년 시기에 쓰기 부진을 판별하기 위한 어휘를 제안하고, 인공지능 지식이 부족한 현장교사도 쉽게 쓰기 결과물 이미지 데이터를 활용하여 쓰기 부진 판별 AI 프로그램의 개발과 적용 가능성을 찾을 수 있었다.
Abstract
This study aims to assess the precision of an AI model crafted for identifying signs of poor writing skills in early elementary students via handwriting image analysis. By examining the handwriting of first-grade students, their writ- ing abilities were categorized into normal and poor, based on selected words that include complex elements such as double consonants, clusters, and diphthongs. The handwriting images collected were then used to develop an artificial intelligence model based on Entry AI block that categorizes learners into two groups according to the characteristics of the vocabulary used. The comparative analysis of the model's accuracy in differentiating between these groups revealed that the model trained with vocabulary featuring a high frequency of clusters achieved the highest accuracy. This find- ing suggests the potential for selecting specific vocabularies that can aid in identifying early writing difficulties of chil- dren and for devoloping AI programs to determine poor writing from the image data of childrens' handwriting results.
참고문헌 (17)
- Kang, D. H., "Achievements and Directions of Responsible Hangul Education Policy for Cultivating Basic Literacy", The Korea Society of Elementary Korean Education, 76, pp.278~303 (2023)
- Development of literacy, numeracy, and diagnostic tools to ensure basic academic ability (Ⅱ), "Park, S.", Korea Institute for Curriculum and Evaluation, ~ (2021)
- Ministry of Education, "The first comprehensive plan to guarantee basic academic ability", ~ (2022)
- Han, M. R., "A Study on the AI Based MNIST Handwriting Recognit", Journal of the Korea IT Policy Management Society, 11(6), pp.1509~1517 (2019)
- Kim, M. H., "A deep learning-based Korean handwriting recognition model for automatic grading of low-grade children's workbook", Database Research, 39(1), pp.85~95 (2023)
- A Study on the Development of KISE-Basic Academic Achievement Tests, "Park, K. S.", Korea Institute For Special Education, 1~386, pp.12 (2004)
- Implementation plan for guaranteeing basic education in 2024, "Chungcheongbuk-do Office of Education", ~ (2024)
- Lee, G., "Exploring the category and level of literacy as a basic academic ability", New Korean Language Education, (126), pp.91~113 (2021)
- Kim, Y. H., "A longitudinal study on prediction of word recognition, reading fluency, and reading comprehension of 4-year-old Children", Institute of Special Education & Rehabitation Science, 54(4), pp.45~70 (2015)
- Lee, Y. E., "The Effects of a Morphological Awareness-based Intervention Program on Spelling for School-aged Children with Spelling Difficulties", Communication Sciences and Disorders, 28(2), pp.255~273 (2020)
- Jang, W., "Writing Skills in 1st to 2nd Graders with Language Difficulties, Typical Development, and Dyslexia with", without Language Difficulties, 2023, pp.28(1), 71~83
- Park, Y., "The Development of Interactive Artificial Intelligence Blocks for Image Classification", Journal of the Korean Association of Information Education, 25(6), pp.1015~1024 (2021)
- Jeong, L., "Development and Application of AI Education Program for Image Recognition for Low Grade Elementary School Students", Journal of the Korean Association of information Education, 26(1), pp.1~10 (2022)
- Kim, B., "Development of Elementary School AI Education Contents using Entry Text Model Learning", Journal of The Korean Association of Information Education, 26(1), pp.65~73 (2022)
- Yang, J. Y., "Extracting characteristics of underachievers learning using artificial intelligenceand researching a prediction model", Journal of the Korea Institute of Information and Communication Engineering, 26(4), pp.510~518 (2022)
- Han, J. H., "AI Classification of Learning Concentration Tendency of Elementary School Students Based on the Rey–Osterrieth Complex Figure Test", Journal of Studies on Schools and Teaching, 8(2), pp.195~214 (2023)
- Jeong, H. S., "A Study on the Adaptive Vocabulary of Korean as a Subject for KSL Elementary School Learners", pp.2022