Predicting pre-click quality for native advertisements
- Submitting institution
-
University of Nottingham, The
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 1324652
- Type
- E - Conference contribution
- DOI
-
10.1145/2872427.2883053
- Title of conference / published proceedings
- WWW '16 Proceedings of the 25th International Conference on World Wide Web
- First page
- 299
- Volume
- 2016-April
- Issue
- -
- ISSN
- -
- Open access status
- -
- Month of publication
- April
- Year of publication
- 2016
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
3
- Research group(s)
-
-
- Citation count
- 11
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The paper proposes a novel learning framework for optimizing advertisement recommendation by incorporating the quality (user experience) of the ads, rather than solely being reliant on the short-term revenue generated. The method provides insightful metrics, features and models for ad recommendation, which enable long-term user engagement. The models were empirically experimented and deployed on Yahoo news streams that include native ads on both mobile and desktop, demonstrating superior user experience for millions of users.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
