FROSTT

The Formidable Repository of Open Sparse Tensors and Tools

Home » Tensors » Amazon Reviews

Amazon Reviews

Product reviews from Amazon and collected by SNAP. The modes represent user-product-word, and each non-zero is the number of times a word appears in a given review. We pre-processed the review text by removing stop words and performing Porter stemming.

Please note that we no longer have the mappings for this dataset. It is provided as a peformance benchmark.

Tensor Statistics

Non-zeros 1,741,809,018
Order 3
Dimensions 4,821,207 x 1,774,269 x 1,805,187
Tags counts , text

Downloadable Files

File Description
amazon-reviews.tns.gz Amazon-Reviews tensor

Citation

@inproceedings{mcauley2013,
  title={Hidden factors and hidden topics: understanding rating dimensions with review text},
  author={McAuley, Julian and Leskovec, Jure},
  booktitle={Proceedings of the 7th ACM conference on Recommender systems},
  pages={165--172},
  year={2013},
  organization={ACM}
}

Discussion