Syntactic And Lexical Comparison Between Ai-generated Reading Passages And Japanese Universities' National Test

This research aimed to develop a methodology for creating mock English reading tests using AI-generated passages to help decrease the workload of EFL teachers who struggle in creating the high-quality mock English reading test for universities’ entrance examinations. To achieve this goal, this pape...

Full description

Saved in:
Bibliographic Details
Main Authors: Satoshi Kurokawa, Maelys Salingre
Format: Article
Language:English
Published: University of Bejaia Abderrahmane Mira 2025-05-01
Series:The Journal of Studies in Language, Culture and Society
Subjects:
Online Access:https://univ-bejaia.dz/revue/jslcs/article/view/590
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850072294644252672
author Satoshi Kurokawa
Maelys Salingre
author_facet Satoshi Kurokawa
Maelys Salingre
author_sort Satoshi Kurokawa
collection DOAJ
description This research aimed to develop a methodology for creating mock English reading tests using AI-generated passages to help decrease the workload of EFL teachers who struggle in creating the high-quality mock English reading test for universities’ entrance examinations. To achieve this goal, this paper examined the lexical and syntactical differences between English Subject of the Center Test at Japanese universities’ entrance exams (ESCT) and AI-generated passages. Three different generative AI were used: OpenAI ChatGPT version 4, Google Gemini version 1.5 Flash, and DeepSeek-V3. To make the vocabulary coverage between AI-generated passages and ESCTs meaningful, topics based on ESCTs were used to create 11 prompts for AI-generated passages. This paper examined text coverage using CEFR-based wordlist and syntactic complexity using the Python library spaCy. Findings revealed that the proportion of A2 level tokens does not differ greatly between AI-generated passages (ChatGPT: 19.1%; Gemini: 17.4%; DeepSeek: 17.3%) and ESCT (15.6%); ESCT had more complex but shorter sentences comparing to AI-generated passages, and personal pronouns accounted for 1.3% of ESCT tokens, while they accounted for less than 1% of generative AIs tokens (ChatGPT, 0.39% ; Gemini, 0.71% ; DeepSeek 0.45%). Few wh-pronouns and the existential there were found in AI-generated passages. This study concluded that when the EFL teachers convert AI-generative text to the mock ESCT, they should (a) add a wider variety of A1 level lemmas, (b) rewrite more complex and shorter sentences, (c) increase personal pronouns and determiners, and (d) reduce adjective modifiers, conjuncts, and coordination.
format Article
id doaj-art-fd5544f06d6d4615ab6184fbfd787225
institution DOAJ
issn 2716-9189
2676-1750
language English
publishDate 2025-05-01
publisher University of Bejaia Abderrahmane Mira
record_format Article
series The Journal of Studies in Language, Culture and Society
spelling doaj-art-fd5544f06d6d4615ab6184fbfd7872252025-08-20T02:47:06ZengUniversity of Bejaia Abderrahmane MiraThe Journal of Studies in Language, Culture and Society2716-91892676-17502025-05-0181Syntactic And Lexical Comparison Between Ai-generated Reading Passages And Japanese Universities' National TestSatoshi Kurokawa0Maelys Salingre1Nagoya UniversityShizuoka University This research aimed to develop a methodology for creating mock English reading tests using AI-generated passages to help decrease the workload of EFL teachers who struggle in creating the high-quality mock English reading test for universities’ entrance examinations. To achieve this goal, this paper examined the lexical and syntactical differences between English Subject of the Center Test at Japanese universities’ entrance exams (ESCT) and AI-generated passages. Three different generative AI were used: OpenAI ChatGPT version 4, Google Gemini version 1.5 Flash, and DeepSeek-V3. To make the vocabulary coverage between AI-generated passages and ESCTs meaningful, topics based on ESCTs were used to create 11 prompts for AI-generated passages. This paper examined text coverage using CEFR-based wordlist and syntactic complexity using the Python library spaCy. Findings revealed that the proportion of A2 level tokens does not differ greatly between AI-generated passages (ChatGPT: 19.1%; Gemini: 17.4%; DeepSeek: 17.3%) and ESCT (15.6%); ESCT had more complex but shorter sentences comparing to AI-generated passages, and personal pronouns accounted for 1.3% of ESCT tokens, while they accounted for less than 1% of generative AIs tokens (ChatGPT, 0.39% ; Gemini, 0.71% ; DeepSeek 0.45%). Few wh-pronouns and the existential there were found in AI-generated passages. This study concluded that when the EFL teachers convert AI-generative text to the mock ESCT, they should (a) add a wider variety of A1 level lemmas, (b) rewrite more complex and shorter sentences, (c) increase personal pronouns and determiners, and (d) reduce adjective modifiers, conjuncts, and coordination. https://univ-bejaia.dz/revue/jslcs/article/view/590Generative AINatural language processinText coverageSyntactic complexityUniversity entrance examination
spellingShingle Satoshi Kurokawa
Maelys Salingre
Syntactic And Lexical Comparison Between Ai-generated Reading Passages And Japanese Universities' National Test
The Journal of Studies in Language, Culture and Society
Generative AI
Natural language processin
Text coverage
Syntactic complexity
University entrance examination
title Syntactic And Lexical Comparison Between Ai-generated Reading Passages And Japanese Universities' National Test
title_full Syntactic And Lexical Comparison Between Ai-generated Reading Passages And Japanese Universities' National Test
title_fullStr Syntactic And Lexical Comparison Between Ai-generated Reading Passages And Japanese Universities' National Test
title_full_unstemmed Syntactic And Lexical Comparison Between Ai-generated Reading Passages And Japanese Universities' National Test
title_short Syntactic And Lexical Comparison Between Ai-generated Reading Passages And Japanese Universities' National Test
title_sort syntactic and lexical comparison between ai generated reading passages and japanese universities national test
topic Generative AI
Natural language processin
Text coverage
Syntactic complexity
University entrance examination
url https://univ-bejaia.dz/revue/jslcs/article/view/590
work_keys_str_mv AT satoshikurokawa syntacticandlexicalcomparisonbetweenaigeneratedreadingpassagesandjapaneseuniversitiesnationaltest
AT maelyssalingre syntacticandlexicalcomparisonbetweenaigeneratedreadingpassagesandjapaneseuniversitiesnationaltest