| 研究課題/領域番号 |
20J13239
|
|---|
| 研究種目 |
特別研究員奨励費
|
| 配分区分 | 補助金 |
|---|
| 応募区分 | 国内 |
|---|
| 審査区分 |
小区分02100:外国語教育関連
|
|---|
| 研究機関 | 東京工業大学 |
|---|
研究代表者 |
PUTRA JAN WIRA GOTAMA 東京工業大学, 情報理工学院, 特別研究員(DC2)
|
|---|
| 研究期間 (年度) |
2020-04-24 – 2022-03-31
|
| 研究課題ステータス |
完了 (2021年度)
|
| 配分額 *注記 |
2,100千円 (直接経費: 2,100千円)
2021年度: 1,000千円 (直接経費: 1,000千円)
2020年度: 1,100千円 (直接経費: 1,100千円)
|
|---|
| キーワード | Argumentative Structure / EFL learners / ICNALE-AS2R Corpus / Annotation / Sentence Reordering / Structure Parsing / EFL students / argumentative structure / annotation tool / TIARA / annotator agreement |
|---|
| 研究開始時の研究の概要 |
This research aims to develop a tool to enable non-native English speakers in writing better argumentative essays (in English). Specifically, it works by arranging sentences in the essay such that it results in a better logically-structured text.
Our main approach is to analyse the argumentative structure. Argumentative structure analysis aims to understand how sentences are connected to each other when forming argument as a whole. The way how to order sentences properly can be inferred from such structure. The tool is useful especially for helping teacher in giving feedback to students.
|
| 研究実績の概要 |
My research aims to develop a support tool for analysis and improving argumentative essays written by English-as-foreign-language (EFL) learners. The tool extracts and visualizes the argumentative structure of a given essay, and then suggests an improved version of the essay by reordering sentences. This project was divided into several steps:
(a) Dataset Construction. In 2020, I constructed the first corpus of 434 essays written by EFL learners (ICNALE-AS2R), which have been annotated with argumentative structure and sentence reordering. I also developed an in-house annotation software (open-source) TIARA for this step.
(b) Structure Parsing Model. Given an input essay, I developed a deep learning system to predict the argumentative structure of the essay. I used the combination of Biaffine Attention and BERT models. I also proposed multi-task and multi-corpora training strategies which enhanced the parsing performance of the base model. The idea was to guide the model on how to reduce the search space and to utilise existing non-EFL datasets via selective sampling.
(c) Sentence Reordering Model. Given an essay and its corresponding argumentative structure as input, I proposed a reordering system based on the linguistic theory of coherence. The task was formulated as a traversal problem, containing two steps. The first is a pairwise ordering constraint task between pairs of sentences (tackled using ALBERT model). The second is a traversal (output generation) step using an ad-hoc algorithm. Detailed empirical evaluation showed that my proposed system has a good potential.
|
| 現在までの達成度 (段落) |
令和3年度が最終年度であるため、記入しない。
|
| 今後の研究の推進方策 |
令和3年度が最終年度であるため、記入しない。
|
