This is a new concept discussion platform introducing AI host. We have developed a discussion platform centered on a fair AI host capable of offensive comment masking and summarizing discussion content.
This repository contains the source code of the Team A of Capstone Design Project at SKKU (Spring 2022).
| name | contribute | responsible folder |
|---|---|---|
| Kim Seok | model serving | made ./server folder |
| Park Jinwoo | summarization model | made ./summarization folder |
| Lee Chungsoo | application dev | made ./Discussion-Korea and ./Discussion-Korea-Backend folder |
| Jang Chaeyoon | offensive comment classifier | made ./classifier folder |
If you want to make your own AI host, please check the ./classifier and ./summarization folder.
./classifier
./multi
./src
├── main.py - main
├── model.py - contain all models
├── train.py - train/eval/test
...
└── utils - utilities
├── dataset.py - dataset & dataloader
├── preprocessing.py - dataset preprocessing & save
...
./binary
...
./crawling
├── main.py - main
...
If you want to check the application code, please check the ./app folder.
[1] Park, Sungjoon, et al. "Klue: Korean language understanding evaluation.", arXiv preprint arXiv:2105.09680, (2021).
[2] Liu, Yinhan, et al. "RoBERTa: A Robustly Optimized BERT Pretraining Approach.", arXiv preprint arXiv:1907.11692, (2019).
[3] Sun, Chi, et al. "How to fine-tune bert for text classification?." China national conference on Chinese computational linguistics., (2019).
[4] Mike Lewis, et al. "Bart: Denoising sequence-tosequence pre-training for natural language generation, translation, and comprehension.", arXiv preprint arXiv:1910.13461, (2019).