Analysis of JLPT N3 Level for Time Expression Sentence Patterns in GPT-5

DOI

10.2991/978-2-38476-563-8_32How to use a DOI?

Keywords

ChatGPT; GPT-5; Syntax; JLPT N3; Time expression sentence patterns

Abstract

Recently, ChatGPT has made various impacts, especially in utilizing foreign language learning. In general, ChatGPT can generate sample sentences in various languages, including Japanese. However, more research is needed to determine the reliability of these responses based on internet data. Therefore, it is important to conduct further verification and human intervention on the answers. This study aims to evaluate the sentences generated by GPT-5, the newest version of ChatGPT, as well as to compare them with example sentences contained in the Japanese language learning textbook Shin Kanzen Master Bunpou N3. The focus of this study is the JLPT N3 level for time expression sentence patterns, namely 「～うちに」, 「～間・～間に」, 「～てからでないと・～てからでなければ」, and 「～ところだ・～ところ（＋助詞）」. This study used a qualitative-comparative analysis method with a descriptive approach. Data will be collected from the responses generated by GPT-5 based on predetermined prompts. Furthermore, the response sentences will be compared with time expression sentence patterns in the textbook and analyzed using a functional approach. Based on the analysis results, the data from this study shows that GPT-5 answers are reasonably accurate, making it a useful recommendation for reference in creating sentences in Japanese. However, ChatGPT should be used in combination with other learning resources to obtain more accurate responses.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Dhaniswari Ananta Ayu
AU  - Intan Khodijatul Kubro
AU  - Rizki Andini
PY  - 2026
DA  - 2026/04/28
TI  - Analysis of JLPT N3 Level for Time Expression Sentence Patterns in GPT-5
BT  - Proceedings of the 4th International Conference of Research on Language 2025 (IROLE 2025)
PB  - Atlantis Press
SP  - 411
EP  - 431
SN  - 2352-5398
UR  - https://doi.org/10.2991/978-2-38476-563-8_32
DO  - 10.2991/978-2-38476-563-8_32
ID  - Ayu2026
ER  -