Evaluating the Quality of Code Comments Generated by Large Language Models for Novice Programmers
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910616510267392 |
|---|---|
| author | Fan, Aysa Xuemo Narayanan, Arun Balajiee Lekshmi Hassany, Mohammad Ke, Jiaze |
| author_facet | Fan, Aysa Xuemo Narayanan, Arun Balajiee Lekshmi Hassany, Mohammad Ke, Jiaze |
| contents | Large Language Models (LLMs) show promise in generating code comments for novice programmers, but their educational effectiveness remains under-evaluated. This study assesses the instructional quality of code comments produced by GPT-4, GPT-3.5-Turbo, and Llama2, compared to expert-developed comments, focusing on their suitability for novices. Analyzing a dataset of ``easy'' level Java solutions from LeetCode, we find that GPT-4 exhibits comparable quality to expert comments in aspects critical for beginners, such as clarity, beginner-friendliness, concept elucidation, and step-by-step guidance. GPT-4 outperforms Llama2 in discussing complexity (chi-square = 11.40, p = 0.001) and is perceived as significantly more supportive for beginners than GPT-3.5 and Llama2 with Mann-Whitney U-statistics = 300.5 and 322.5, p = 0.0017 and 0.0003). This study highlights the potential of LLMs for generating code comments tailored to novice programmers. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_14368 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Evaluating the Quality of Code Comments Generated by Large Language Models for Novice Programmers Fan, Aysa Xuemo Narayanan, Arun Balajiee Lekshmi Hassany, Mohammad Ke, Jiaze Software Engineering Artificial Intelligence Human-Computer Interaction Large Language Models (LLMs) show promise in generating code comments for novice programmers, but their educational effectiveness remains under-evaluated. This study assesses the instructional quality of code comments produced by GPT-4, GPT-3.5-Turbo, and Llama2, compared to expert-developed comments, focusing on their suitability for novices. Analyzing a dataset of ``easy'' level Java solutions from LeetCode, we find that GPT-4 exhibits comparable quality to expert comments in aspects critical for beginners, such as clarity, beginner-friendliness, concept elucidation, and step-by-step guidance. GPT-4 outperforms Llama2 in discussing complexity (chi-square = 11.40, p = 0.001) and is perceived as significantly more supportive for beginners than GPT-3.5 and Llama2 with Mann-Whitney U-statistics = 300.5 and 322.5, p = 0.0017 and 0.0003). This study highlights the potential of LLMs for generating code comments tailored to novice programmers. |
| title | Evaluating the Quality of Code Comments Generated by Large Language Models for Novice Programmers |
| topic | Software Engineering Artificial Intelligence Human-Computer Interaction |
| url | https://arxiv.org/abs/2409.14368 |