Dynamic Genre-Specific Automated Online Feedback Tool

Research Project

Project/Area Number	23K00656
Research Category	Grant-in-Aid for Scientific Research (C)
Allocation Type	Multi-year Fund
Section	一般
Review Section	Basic Section 02100:Foreign language education-related
Research Institution	The University of Aizu
Principal Investigator	CARR Nicholas 会津大学, コンピュータ理工学部, 准教授 (70965699)
Co-Investigator(Kenkyū-buntansha)	Julian Villegas 会津大学, コンピュータ理工学部, 上級准教授 (50706281) BLAKE John 会津大学, コンピュータ理工学部, 上級准教授 (80635954)
Project Period (FY)	2023-04-01 – 2026-03-31
Project Status	Granted (Fiscal Year 2023)
Budget Amount *help	¥3,640,000 (Direct Cost: ¥2,800,000、Indirect Cost: ¥840,000) Fiscal Year 2025: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000) Fiscal Year 2024: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000) Fiscal Year 2023: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Keywords	Automative feedback / Dynamic assessment / Sociocultural theory / iCALL / L2 Writing / Automated feedback / Dynamic feedback / Computer-science writing
Outline of Research at the Start	This project develops an online tool which identifies genre specific errors in computer-science academic writing; errors which are not detected in generic grammar checking tools. The tool then provides dynamic feedback to users. The learning benefits of the tool are investigated longitudinally.
Outline of Annual Research Achievements	The Dynamic Genre-Specific Automated Online Feedback Tool we are developing is important because there is no online automated corrective feedback tool that provides dynamic feedback on the linguistic features of computer science writing. With the release of OpenAI, we are pursuing the incorporation of added features to our original tool - mainly targeting grammar errors. A thorough search of the literature shows that automated dynamic assessment of additional language writing is challenging due to not being able to predict learner errors. However, we are hoping to develop a tool which, through the use of generative AI, can take a step towards an interactionist type of automated dynamic assessment. The protoype is almost finished and multi-modal feedback explanations have been created.
Current Status of Research Progress	Current Status of Research Progress 3: Progress in research has been slightly delayed. Reason We are making good progress. However, due to the release of OpenAI and other generative AI after applying for this finding, we are investigating the expansion of the tool to not only target genre-specific errors of computer science but also grammatical errors. This addition has meant we are very slighty behind our original schedule of releasing the prototype version of the tool.
Strategy for Future Research Activity	We intend to spend some more time further exploring the possibility of using generative AI for our tool. Such investigation is important because it will allow our tool to provide greater assistance to users. Once this is complete, the prototype version of the tool will be available for students to use and we will conduct multiple rounds of usability tests to refine the tool this academic year. This will set up the primary project, which is investigate the learning effects of using the tool over a full semester in the following year. We are always planning on submitting articles/book chapters to publish the development phase of the tool later in 2024.