Accurately interpreting clickthrough data as implicit feedback

By Thorsten Joachims, Laura A. Granka, Bing Pan, Helene Hembrooke, Geri Gay. 2005

1,564 citations recorded by the source corpus.

Subfields: Information Systems, Information Retrieval & Search

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Information Systems

In Information Retrieval & Search

Citations in the corpus

Learning to summarize from human feedback (2020, background): “A human judges which is a better summary of the post.”
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].”
Boosting Search Engines with Interactive Agents (2021, neutral): “Web searchers expect the best answer to be among the top few hits on the ﬁrst results page (Hearst, 2009, §5) and pay marginal attention to the bottom half of the 10 blue links (Granka et al., 2004; Joachims et al., 2005; Nielsen & Pernice, 2009; Strzelecki, 2020).”
Monotonic Calibrated Interpolated Look-Up Tables (2015, methodology): “Also, using bottom-clicked pairs removes the trust bias that users know they are being presented with a ranked list and prefer samples that are ranked-higher (Joachims et al., 2005).”
Understanding web browsing behaviors through Weibull analysis of dwell time (2010, background): “, [12, 13, 24]), and the other based on large-scale log analysis (e.”
The EMPATHIC Framework for Task Learning from Implicit Human Feedback (2020, background): “Others’ usage of “implicit feedback” has referred to the implied feedback when a human refrains from giving explicit feedback [30, 31], to human biomagnetic signals [32], or to facial expressions [33, 34, 35].”
User performance versus precision measures for simple search tasks (2006, background): “Such a search task may be expected to more closely reﬂect the system eﬀectiveness that metrics such as MAP and bpref are measuring: recent research has demonstrated that users focus on the top ranked answers when looking at a ranked list of search results [15].”
Query chains: learning to rank from implicit feedback (2005, background): “Further analysis showed that this is usually the abstract immediately below the one clicked on [17].”
Efficient multiple-click models in web search (2009, background): “Previous eye-tracking experiments and studies on explaining position-bias of user clicks provide a spectrum of hypotheses and models on how an average user examines and possibly clicks web documents returned by a search engine with respect to the submitted query.”
Situational Context for Ranking in Personal Search (2017, methodology): “In our ﬁnal ranking stage, we consider the position bias problem [26] similar to what is described in [44].”
Adversarial Web Search (2011, background): “The fact that certain areas of Web pages tend to be clicked more often independently of their relevance to the user task has been observed for several types of Web pages (not only search result pages), and it affects the interpretation of Web clicks in general [115].”
Label ranking by learning pairwise preferences (2008, background): “This information can be turned into binary preferences by assuming that the selected pages are preferred over nearby pages that are not clicked on (Joachims et al., 2005).”
Understanding PubMed® user search behavior through log analysis (2009, result): “Our finding comports with studies that show similar patterns in eye movements and mouse movements during Web search (27,28,77).”
What are you looking for?: an eye-tracking study of information usage in web search (2007, methodology): “[12] is a possible exception to this because they did manipulate the order of search results for some of their users.”
Tailoring click models to user goals (2009, background): “Previous eyetracking studies [9, 10] indicated that clicks are biased as a form of absolute relevance judgement, and clicking decision on a web document depends on both the position (rank) and the context (other documents) of the presentation.”
Information Retrieval: A Biomedical and Health Perspective (2020, background): “Although peer review had its origins centuries ago, it did not achieve its present form until the twentieth century [146].”