Datasets:

Modalities:
Text
Formats:
parquet
Languages:
Azerbaijani
DOI:
Libraries:
Datasets
Dask
License:

Source code

#1
by jafarisbarov - opened

Are you planning to share the source code for this?

LocalDoc org

Are you planning to share the source code for this?

Hi. Scraper source code or texts in raw ?

Scraping and text processing scripts. But I also think it would also be useful to have metadata (titles, urls, etc.) in the database.

LocalDoc org

Scraping and text processing scripts. But I also think it would also be useful to have metadata (titles, urls, etc.) in the database.

I don't plan to put the data mining project in the public for now, because I'm expanding it for a big project. As for metadata, I decided to do without it this time. But you can separate the title from the text. The first sentence of the article is the title. In future projects, I will take this observation into account and try to always include metadata in the dataset

Thanks for the info (and the dataset!)

vrashad changed discussion status to closed

Sign up or log in to comment