You need to agree to share your contact information to access this dataset

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this dataset content.

Data sources come from the following categories: 1.Web crawler dataset:

  • Website UET (ĐH Công nghệ): tuyensinh.uet.vnu.edu.vn; new.uet.vnu.edu.vn
  • Website HUS (ĐH KHTN): hus.vnu.edu.vn
  • Website EUB (ĐH Kinh tế): ueb.vnu.edu.vn
  • Website IS (ĐH Quốc tế): is.vnu.edu.vn
  • Website Eduacation (ĐH Giáo dục): education.vnu.edu.vn
  • Website NXB ĐHQG: press.vnu.edu.vn
    List domain web crawler
  1. CC100:
    link to CC100 vi
  2. Vietnews:
    link to bk vietnews dataset
  3. C4_vi:
    link to C4_vi

Folder Toxic store files demo for toxic filtering.We filtered C4_validation dataset, vietnews samples dataset and a part(1/50) of CC100_vi dataset.After the process, datasets split into nontoxic part and toxic part.

Folder Dedup store files after the deduplication process of above files.
Folder Toxic_2, Dedup_2, Tokenized_2 is the result of the second process that we executed on 17 files of C4_vi dataset wich contains 1b tokens

Downloads last month
78