|
--- |
|
license: apache-2.0 |
|
--- |
|
data source：https://cseweb.ucsd.edu/~jmcauley/datasets/amazon_v2/ |
|
|
|
we construct a new dataset Amazon reviews (Ni et al., 2019) on data aggregated over six genres 5core: beauty, fashion, appliances, giftcards, magazines, and software. |
|
|
|
cite: |
|
Jianmo Ni, Jiacheng Li, and Julian McAuley. Justifying recommendations using distantly-labeled |
|
reviews and fine-grained aspects. In Empirical Methods in Natural Language Processing and |
|
International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), 2019. URL |
|
https://www.aclweb.org/anthology/D19-1018. |
