abrhaleitela/Sentiment-Analysis-for-Tigrinya

First sentiment analysis (with two classes) dataset for Tigrinya language. This train set is constructed automatically while the test set labeled manually. Negative sentiment labeled as 0. Positive sentiment labeled as 1. train size: 49374 (25031 negative and 24343 positive) test size: 4000 (2000 negative and 2000 positive) For more information on our experiments and results please check our paper: Title: Transferring Monolingual Model to Low-Resource Language: The Case of Tigrinya