Beyond position bias: examining result attractiveness as a source of presentation bias in clickthrough data
By Yisong Yue, Rajan Patel, Hein Roehrig. 2010
223 citations recorded by the source corpus.
Subfields: Computer Science Applications, 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 Computer Science Applications
- Kristina Lerman
- Thorsten Joachims
- Tobias Schnabel
- Tad Hogg
- Adith Swaminathan
- Xuanhui Wang
- Donald Metzler
- Marc Najork
In Information Systems
- Thorsten Joachims
- Michael Bendersky
- Adith Swaminathan
- Tobias Schnabel
- Xuanhui Wang
- Kristina Lerman
- Donald Metzler
- Marc Najork
Citations in the corpus
- Leveraging Position Bias to Improve Peer Recommendation (2014, background): “For instance, presentation order affects which items in a list of search results users click on [8–10], and the answer they select when responding to a multiple choice question [6,11].”
- Unbiased Learning-to-Rank with Biased Feedback (2016, background): “Furthermore, it is possible to model saliency biases [31] by replacing the pr with a regression function.”
- Beyond Ranking: Optimizing Whole-Page Presentation (2016, background): “Studies observe more interesting user interaction patterns, notably vertical bias [12, 26, 31] and presentation bias [39].”
- Bias in algorithmic filtering and personalization (2013, background): “Yue et al. (2010) report that the attractiveness of information can also cause presentation bias if the title and abstract of a resource is bolded, it generates more clicks.”
- Time-Aware Click Model (2016, background): “DOI: http://dx.doi.org/10.1145/2988230”
- Up next: retrieval methods for large scale related video suggestion (2014, background): “However, research shows that the click-through rate can be highly biased by factors such as position and attractiveness of the presentation [32].”
- Learning to Rank with Selection Bias in Personal Search (2016, methodology): “As a special case, when n = 2, the randomization reduces to the previously proposed FairPairs algorithm [11, 33].”
- Position Bias Estimation for Unbiased Learning to Rank in Personal Search (2018, background): “However, a well-known challenge in learning from click data is its inherent bias: position bias [19], presentation bias [32], and trust bias [26], etc.”
- Beliefs and biases in web search (2013, background): “Yue and colleagues [47] studied the effect of caption attractiveness, defined for their study as the presence and absence of bolded terms in the titles and snippets of the caption.”