This sentiment dataset was used in the paper: John Blitzer, Mark Dredze, Fernando Pereira. Biographies, Bollywood, Boom-boxes and Blenders: Domain Adaptation for Sentiment Classification. Association of Computational Linguistics (ACL), 2007.
The author asks, if you use this data for your research or a publication, to cite the above paper as the reference for the data, and to inform him about the reuse.
The Multi-Domain Sentiment Dataset contains product reviews taken from Amazon.com from 4… See the full description on the dataset page:
https://huggingface.co/datasets/katossky/multi-domain-sentiment.