electra-base-cyberbullying

This is an ELECTRA Small model for the Japanese language finetuned for automatic cyberbullying detection.

The model was based on Izumi Lab ELECTRA small Japanese discriminator, and later finetuned on a balanced dataset created by unifying two datasets, namely "Harmful BBS Japanese comments dataset" and "Twitter Japanese cyberbullying dataset".

Licenses

The finetuned model with all attached files is licensed under CC BY-SA 4.0, or Creative Commons Attribution-ShareAlike 4.0 International License.

Citations

Please, cite this model using the following citation.

@inproceedings{tanabe2022electra-small-cyberbullying,
  title={北見工業大学 テキスト情報処理研究室 ELECTRA Small ネットいじめ検出モデル (Izumi Lab ver.)}, 
  author={田邊　威裕　and プタシンスキ ミハウ and エロネン ユーソ and 桝井 文人}, 
  publisher={HuggingFace}, 
  year={2022},
  url = "https://huggingface.co/kit-nlp/electra-small-japanese-discriminator-cyberbullying"
}