Natural language generation of trend descriptions for pedagogic purposes
| Project/Area Number |
22K00792
|
|---|
| 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 |
BLAKE John 会津大学, コンピュータ理工学部, 上級准教授 (80635954)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
Pyshkin Evgeny 会津大学, コンピュータ理工学部, 上級准教授 (50794088)
|
|---|
| Project Period (FY) |
2022-04-01 – 2025-03-31
|
| Project Status |
Granted (Fiscal Year 2022)
|
| Budget Amount *help |
¥3,770,000 (Direct Cost: ¥2,900,000、Indirect Cost: ¥870,000)
Fiscal Year 2024: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2023: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2022: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
|
|---|
| Keywords | NLG / trend descriptions / describing graphs / intelligent CALL / data series description / NLP / language generation |
|---|
| Outline of Research at the Start |
This interactive online tool provides unlimited practice opportunities to describe graphs and charts. Students can practice at three levels: word, clause or sentence using generated practice texts. Students either fill in the gaps, complete sentence stems or draft the whole text. On completion of their practice task, they compare their answers with an automatically generated plain or colorized exemplar text. This helps learners notice patterns, which is said to be a precursor to learning.
|
| Outline of Annual Research Achievements |
A corpus of trend descriptions was compiled and manually analyzed to identify rhetorical patterns. The initial codebase of the description generator was created. We are still working on enabling examplars and practice descriptions to be generated at three different difficulty levels. An automated workflow incorporating AI-generated video explanations that are moderated by language experts has been trialled. A preliminary set of explanation videos has been created for piloting purposes.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
We have been able to follow our timeline set at the outset. The codebase will continue to be improved throughout the project, but the fundamental design and the core functions has been completed.
|
| Strategy for Future Research Activity |
A more sophisticated combination of rule-based and probabilistic parsing will be used to create a powerful NLG pipeline. This pipeline will draw on published research to provide users with access to cutting-edge research in NLG. At the end of this year, we expect to deploy an online version for testing purposes.
|
Report
(1 results)
Research Products
(6 results)
