Algorithmic Learning: Assessing the Potential of Large Language Models (LLMs) for Automated Exercise Generation and Grading in Educational Settings

被引：0

作者：

Zou, Wenlong ^{[1
]}

Goh, Tiong-Thye ^{[2
]}

Zhu, Huiting ^{[1
]}

Liu, Mengjun ^{[1
]}

Yang, Bing ^{[1
]}

机构：

[1] Hubei Univ, Sch Comp Sci, Youyi Rd 368, Wuhan 430062, Hubei, Peoples R China

[2] Victoria Univ Wellington, Sch Informat Management, Wellington, New Zealand

来源：

INTERNATIONAL JOURNAL OF HUMAN-COMPUTER INTERACTION | 2025年

基金：

中国国家自然科学基金;

关键词：

ChatGPT; algorithmic programming; large language models; evaluation;

D O I：

10.1080/10447318.2025.2520931

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

This study explores ChatGPT's role as a teaching aid in algorithmic education, focusing on its ability to generate and evaluate algorithmic questions and solutions. Qualitative analysis shows strong performance in Sensibleness, Readiness, and Topicality, though Novelty remains an area for improvement. ChatGPT also demonstrated self-improvement in criteria like Efficiency and Robustness, with over 60% enhancement. A comparison of AI and teacher grading revealed that ChatGPT provided accurate assessments, closely aligning with expert evaluations. The findings highlight ChatGPT's potential in educational assessment and call for further exploration of student perceptions and ethical considerations.

引用

页数：18

共 39 条

[1] A Feasibility Study on Automated SQL Exercise Generation with ChatGPT-3.5 [J].

Aerts, Willem ;

Fletcher, George ;

Miedema, Daphne .

PROCEEDINGS OF THE 3RD ACM SIGMOD INTERNATIONAL WORKSHOP ON DATA SYSTEMS EDUCATION: BRIDGING EDUCATION PRACTICE WITH EDUCATION RESEARCH, DATAED 2024, 2024, :13-19

[2]

Ausat A. M. A., 2023, Journal on Education, V5, P16100, DOI [10.31004/joe.v5i4.2745, DOI 10.31004/JOE.V5I4.2745]

[3]

Baidoo-Anu D., 2023, Journal of AI, V7, P52, DOI [DOI 10.2139/SSRN.4337484, 10.61969/jai.1337500, DOI 10.61969/JAI.1337500]

[4]

Black P. E., 2005, Dictionary of algorithms and data structures, V2, P62, DOI [DOI 10.18434/T4/1422485, 10.18434/T4/1422485]

[5]

Brandao LD, 2012, PROC FRONT EDUC CONF

[6] ChatGPT in a programming course: benefits and limitations [J].

Bringula, Rex .

FRONTIERS IN EDUCATION, 2024, 9

[7]

Brown TB, 2020, ADV NEUR IN, V33

[8]

Chen Mark, 2021, Evaluating Large Language Models Trained on Code, DOI DOI 10.48550/ARXIV.2107.03374

[9]

Cormen T.H., 2022, Introduction to Algorithms

[10]

da Costa RCAB, 2020, PROC FRONT EDUC CONF

← 1 2 3 4 →