TasaractDavut Emre Taşar
Go back
Model

Turkish Sentiment Analysis · TSAM

Wetin e dey do

Give di model one sentence for Turkish, and e go score three ways wey person fit read di tone. For di settings wey dem publish, 0 mean negative, 1 mean neutral and 2 mean positive. For di result, di name of di neutral group na “Notr”. If you dey count reviews, you fit count di neutral comments on their own.

One model, four sets of Turkish data

One independent study compare four models with four sets of Turkish data. For di test wey get two classes, di researchers remove neutral labels. Dis model get F1 score of 0.97 for Humir and 0.54 for Twt.

12,053Total downloads for di Hub18 Nọv 2026Hugging Face Hub API

One earlier model wey I train for reviews

I train Turkish BERT make e fit tell whether person write positive or negative review. I put di training settings for GitHub.

Original documents

Original source · huggingface.co
How to cite am

Turkish Sentiment Analysis · TSAM · https://huggingface.co/emre/turkish-sentiment-analysis