Beyond Ranking: Optimizing Whole-Page Presentation
By Yue Wang, Dawei Yin, Luo Jie, Pengyuan Wang, Makoto Yamada, Yi Chang, Qiaozhu Mei. 2016
86 citations recorded by the source corpus.
Subfields: Artificial Intelligence, Information Systems
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Thorsten Joachims
- Adith Swaminathan
- Alekh Agarwal
- J. Langford
- Akshay Krishnamurthy
- W. Bruce Croft
- Nathan Kallus
- Miroslav Dud'ik
In Information Systems
- W. Bruce Croft
- Thorsten Joachims
- Yongfeng Zhang
- Adith Swaminathan
- Kenta Takatsu
- Ivan Zaitsev
- Aman Agarwal
- Liu Yang
Citations in the corpus
- Off-policy evaluation for slate recommendation (2016, background): “Previous work on off-policy evaluation and whole-page optimization improves the probability of match between logging and evaluation by restricting attention to small slate spaces [35, 26], introducing assumptions that allow for partial matches between the proposed and observed slates [27], or assumi”
- A General Framework for Counterfactual Learning-to-Rank (2018, methodology): “tual approach uses inverse propensity scoring (IPS), originally employed in survey sampling [8] and causal inference from observational studies [17], but more recently also in whole page optimization [27], IR evaluation with manual judgments [18], and recommender evaluation [12, 19]. This approach i”
- Unbiased Learning-to-Rank with Biased Feedback (2016, methodology): “Our approach uses inverse propensity scoring (IPS), originally employed in survey sampling [7] and causal inference from observational studies [19], but more recently also in whole page optimization [29], IR evaluation with manual judgments [20], and recommender evaluation [13, 21].”
- Answer Interaction in Non-factoid Question Answering Systems (2019, background): “ally need to click on the ranked results and be redirected to the documents. Modern search engines have paid attention to search results diversification [3, 21] and heterogeneous content presentation [29]. Recently, several works have focused on retrieving extractive answers instead of documents [11”