The Adressa Dataset is a news dataset that includes news articles (in Norwegian) in connection with anonymized users. We hope that this dataset will be helpful to achieve a better understanding of the news articles in conjunction with their readers.

This dataset is published with the collaboration of Norwegian University of Science and Technology (NTNU) and Adressavisen (local newspaper in Trondheim, Norway) as a part of RecTech project on recommendation technology. For further details of the project and the dataset please refer to the paper mentioned below for citations.

Datasets

• Light version 1: 1.4 GB (compressed) - 1 Week of data collection - 923 articles (in Norwegian), 15.514 users, average article length is 518.6 words
• Light version 2: 16 GB (compressed) - 10 Weeks of data collection

Please note that the full version of the dataset (which is not available on this page) includes more attributes than the light version. The details are explained in the documentation below. If you would like to get the full version or the content of news articles (in Norwegian) please contact us.

Documentation

• Description of the dataset fields and other details about the dataset


• Licensing information: This dataset is available only to use for non-commercial puropses.

Citation

If you use the SmartMedia Adressa Dataset, please cite the following paper:
Gulla, J. A., Zhang, L., Liu, P., Özgöbek, Ö., & Su, X. (2017, August). The Adressa dataset for news recommendation. In Proceedings of the International Conference on Web Intelligence (pp. 1042-1048). ACM.

About Us

SmartMedia Program @ Norwegian University of Science and Technology

Contact Us

reclab (at) idi.ntnu.no

Researching on news related domain?

Share your experiences with the domain experts at:
The International Workshop on News Recommendation and Analytics (INRA)