Create dataset loader for ara-close-lange

[SamuelCahyawijaya]

Dataloader name: ara_close/ara_close.py
DataCatalogue: http://seacrowd.github.io/seacrowd-catalogue/card.html?ara_close

Dataset	ara_close
Description	The dataset contribution of this study is a compilation of short fictional stories written in Bikol for readability assessment. The data was combined other collected Philippine language corpora, such as Tagalog and Cebuano. The data from these languages are all distributed across the Philippine elementary system's first three grade levels (L1, L2, L3). We sourced this dataset from Let's Read Asia (LRA), Bloom Library, Department of Education, and Adarna House.
Subsets	-
Languages	bcl, tgl, ceb
Tasks	Readibility Assessment
License	Creative Commons Attribution 4.0 (cc-by-4.0)
Homepage	https://github.com/imperialite/ara-close-lang
HF URL	-
Paper URL	https://aclanthology.org/2023.findings-acl.331/

[androstj]

#self-assign

[imperialite]

Hello. I'm the owner of this dataset and was about to submit it through the form provided, but it seems someone has already done it. Can I submit a new form? I'm sure I can fill up some of the details more clearly. Also, so I can be attributed for the points.

[androstj]

@imperialite Hi, I un-assign myself, you can take this task :)

[sabilmakbar]

Hi @imperialite do you want to take this dataloader task so you can get points from it?

[sabilmakbar]

Hello. I'm the owner of this dataset and was about to submit it through the form provided, but it seems someone has already done it. Can I submit a new form? I'm sure I can fill up some of the details more clearly. Also, so I can be attributed for the points.

An update to this. You may resubmit it again w/ more complete information. If it's approved by the Datasheet Reviewers, you'll be attributed w/ points too, @imperialite

[sabilmakbar]

Hi @androstj, since he isn't responding to your last comment on this Issue after 2 weeks, I think you may retake this issue again

[imperialite]

Hi @sabilmakbar thanks for this. I'll resubmit the data details to the form again. I'm contributing by adding more data than implementing data loaders. @androstj feel free to retake the issue again and sorry for the delay :)

[SamuelCahyawijaya]

@imperialite: Oh, I just realized that the previous one was about the datasheet. I will recheck the new submission and the score will be shared in this case.

[MJonibek]

#self-assign

[MJonibek]

Can you please advise, which schema I need to use for this dataset?

I was thinking about text schema (for text classification) with labels L1, L2 and L3, but I am not sure.

[imperialite]

Hi @MJonibek yes using the labels L1, L2, and L3 is correct and these are the same as what we used in the paper. Thanks.

[sabilmakbar]

for the schema in SEACrowd, you can use the text label schema (under utils/schema/text.py). However, since the READABILITY_ASSESSMENT task isn't available on the _SUPPORTED_TASK list yet, I'll help with adding that.

[MJonibek]

I found out that data for tgl was deleted.

Also, data has titles and text. For text label schema (utils/schema/text.py),
{
"id": idx,
"text": text,
"label": label,
}

need I:

1. append title to text
2. use title as id
3. just ignore title?

[imperialite]

I deleted the Tagalog (tgl) data because the owner (Adarna House) would prefer if it is requested directly through them. The title can be the ID since it can be used for checking duplication and additional info on the books (useful for genre-based or author analysis).