HF dataset

kanhatakeyama / japanese-corpus-categorized

A cleaned Japanese web corpus (e.g., mc4‑ja) clustered into roughly 10 000 groups by unsupervised learning, usable for lawful analysis, with only some files in Parquet format listed in the “out” folder and downloadable via Git LFS.

Description in other languages

Statistics

Statistics as of Nov 16, 2026

Hugging Face downloads (last 30 days)
620
Hugging Face likes
3
Listed
Oct 29, 2024

Downloads (30-day window) over time

0200400600800NovSepJulMayMarJan 2026620
Show as a table
MonthDL (30d)
Dec 2025442
Jan 2026—
Feb 2026401
Mar 2026159
Apr 2026126
May 2026246
Jun 2026271
Jul 2026209
Aug 2026270
Sep 2026225
Oct 2026326
Nov 2026620

Downloads over the 30 days before the last snapshot of each month.

Popular in the same category

Resources in similar research fields

Learn from slides & articles