Sentiment analysis data and word embeddings for Erzya, Komi-Zyrian, Moksha and Udmurt

The aligned sentiment annotated data is in setiment_eval_data.json, vectors.zip has the word embeddings in a textual Gensim format, code.zip has the code and models.zip the sentiment analysis model.

Please cite the following paper:

Alnajjar, K., Hämäläinen, M., & Rueter, J, (2023) Sentiment Analysis Using Aligned Word Embeddings for Uralic Languages. In Proceedings of the Second Workshop on Resources and Representations for Under-resourced Languages and Domains (RESOURCEFUL-2023)