Home

Awesome

TIARA v2.0: An Interactive Tool for Annotating Discourse Structure and Text Improvement

HomepageLive DemoPaper

License: MIT version

Notice

For people without programming background, it is recommended to visit the homepage instead of this page.

About

TIARA is a tool for annotating argumentative structure and sentence reordering. This is a client-side tool, so you do not need to install anything. You just need to download the TIARA package (click code then download zip on top of this page), and the tool is ready to use; as easy as that. Beyond as an annotation tool, we also designed the tool to be useful for educational purposes as well. It can be used both in learning-to-read and learning-to-write scenarios.

Concept

There are many great annotation tools available out there. However, some (if not many) require complicated steps (and dependencies) to install. Hence, people without a programming background may find it difficult to use them. TIARA considers the ease of use in mind (of course, with compromises), even for people without a programming background. That is why we provide this client-side annotation tool. The tool is customisable; you can configure it to suit your annotation project by changing the conguration file: js/annotation-globalsetting.js. Please read the manual (in the manual/ folder) on how to do this.

We try to design the tool to be as generic as possible, that is, supporting a wide range of tasks. It can be used to annotate (1) tree-shaped discourse structure (2) tree-shaped argumentative structure, (3) reordering sentences, and (4) content editing. Aspect (3) and (4) make this tool unique compared to existing tools. If you are a teacher, you can use this tool to analyse students' writing. You may then provide feedback, telling them where to improve their texts. Students can directly revise their texts in TIARA. One of our team members came from an educational background and has used TIARA to analyse argumentative texts and encourage revisions. Please read our paper below and the manual for a more detailed explanation.

Please do not hestitate to contact us if you have any questions

How to Use

Formatting Text for Annotation

Format the text you want to annotate in .txt, in which each discourse unit (sentence/clause) is separated by a newline. See the following example (sample_original/ESSAY_TRIAL_00.txt).

I agree with the previous statement.
I have a two reasons.
If somebody smokes in the restaurant, other people may not be able to enjoy the experience.
However, if we ban smoking in restaurants, then those restaurants might lose some customers.
But, I firmly support banning smoking in restaurants because we need to prioritise health.
Some restaurants are indeed popular, especially among old men, because they allow people to smoke.
In conclusion, I encourage banning smoking in all restaurants.

Important Notes

We assume you have read the manual at this point.

Paper

The paper describing the tool has been published at LREC2020. Please kindly cite the following paper when you use TIARA. You can use this tool for free, but it would be appreciated if you could send us a courtesy email so we could survey what kind of tasks the tool is used for. Please visit this link to contact us.

@inproceedings{putra-etal-2020-tiara,
    title = "{TIARA}: A Tool for Annotating Discourse Relations and Sentence Reordering",
    author = "Putra, Jan Wira Gotama  and
      Teufel, Simone  and
      Matsumura, Kana  and
      Tokunaga, Takenobu",
    booktitle = "Proceedings of The 12th Language Resources and Evaluation Conference",
    month = may,
    year = "2020",
    address = "Marseille, France",
    publisher = "European Language Resources Association",
    url = "https://www.aclweb.org/anthology/2020.lrec-1.854",
    pages = "6912--6920",
    language = "English",
    ISBN = "979-10-95546-34-4",
}

Jan Wira Gotama Putra, Simone Teufel, Kana Matsumura, and Takenobu Tokunaga. TIARA: A Tool for Annotating Discourse Relations and Sentence Reordering. Proceedings of the 12th International Conference on Language Resources and Evaluation (LREC), pp. 6914--6922, Marseille, France, May 2020.

What can you get from the paper

Update!!

Since we published the LREC paper above, we have made the following updates to the tool.

License

MIT

Dependencies (included)

Screenshots