Preprocess for Text Classification

Question

Preprocess for Text Classification

kentwhf opened this issue 4 years ago · 2 comments

Hi,

I am interested in applying SGC to some other text classification datasets. How did you prepossess the dataset?

Many thanks ahead

Answer 1 · 2021-04-13T17:17:59.000Z

Also, is it possible to release the original text data in .txt format (eg. R8.txt)? I might want try some sensitivity analysis on embedding size as well

Answer 2 · 2021-04-13T20:10:48.000Z

Hi @kentwhf,
Thank you for your interest. Our text classification code is a modified clone of the TextGCN repo. Unfortunately, we simply use their preprocessed data, so we don't have the details of how they preprocess the data. Please contact the author of TextGCN for the data preprocessing.