• Stars
    star
    172
  • Rank 221,201 (Top 5 %)
  • Language
    Shell
  • License
    BSD 3-Clause "New...
  • Created about 7 years ago
  • Updated about 6 years ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

Which Encoding is the Best for Text Classification in Chinese, English, Japanese and Korean?

Glyph

This repository is used to publish all the code used for the following article:

Xiang Zhang, Yann LeCun, Which Encoding is the Best for Text Classification in Chinese, English, Japanese and Korean?, arXiv 1708.02657

The code and datasets are completely released as of January 2018, including all the code for crawling, preprocessing and training on the datasets. However, the documentation may not be complete yet. That said, readers could refer to the doc directory for an example in reproducing all the results for the Dianping dataset, and extend that to other datasets in similar ways.

Reproducibility Manifesto

If anyone sees a number in our paper, there is a script one can execute to reproduce it. No responsibility should be imposed on the user to figure out any experimental parameter barried in the paper's content.

Datasets

The data directory contains the preprocessing scripts for all the datasets used in the paper. These datasets are released separately of their processing source code. See below for details.

Summary

The following table is a summary of the datasets. Most of them have millions of samples for training.

Dataset Language Classes Train Test
Dianping Chinese 2 2,000,000 500,000
JD full Chinese 5 3,000,000 250,000
JD binary Chinese 2 4,000,000 360,000
Rakuten full Japanese 5 4,000,000 500,000
Rakuten binary Japanese 2 3,400,000 400,000
11st full Korean 5 750,000 100,000
11st binary Korean 2 4,000,000 400,000
Amazon full English 5 3,000,000 650,000
Amazon binary English 2 3,600,000 400,000
Ifeng Chinese 5 800,000 50,000
Chinanews Chinese 7 1,400,000 112,000
NYTimes English 7 1,400,000 105,000
Joint full Multilingual 5 10,750,000 1,500,000
Joint binary Multilingual 2 15,000,000 1,560,000

Download

Datasets are released separtely of the source code via links from Google Drive. These datasets should only be used for the purpose of research.

Dataset Train Test
Dianping Link Link
JD full Link Link
JD binary Link Link
Rakuten full Link Link
Rakuten binary Link Link
11st full Link Link
11st binary Link Link
Amazon full Link Link
Amazon binary Link Link
Ifeng Link Link
Chinanews Link Link
NYTimes Link Link
Joint full Link Link
Joint binary Link Link

GNU Unifont

The glyphnet scripts require the GNU Unifont character images to run. The file unifont-8.0.01.t7b.xz can be downloaded via this link.