Evaluating implicit measures to improve web search

By Steve Fox, Kuldeep Karnawat, Mark Mydland, Susan T. Dumais, Thomas White. TOIS, 2005

633 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Information Systems

View Paper: https://www.semanticscholar.org/paper/f194f83d9e137ffddf2fd0e9dd95cdfe6ab3deb4

How researchers see this paper

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

In Economics and Econometrics

In Information Systems

Citations in the corpus

Analyzing and evaluating query reformulation strategies in web search logs (2009, background): “Several later studies agree that click data are indicators of search result preferences and discuss best methods of analyzing click data [2][12].”
Modeling the impact of short- and long-term behavior on search personalization (2012, methodology): “Hence we obtained these judgments using a logbased methodology inspired by [12] and similar to that used in [2][8].”
Test Collection Based Evaluation of Information Retrieval Systems (2010, background): “[87] showed that it was possible to use clicks as indicators of relevance.”
Potential for personalization (2010, background): “[Agchtein et al. 2006; Fox et al. 2005] developed more complex learned models to predict relevance judgments or preferences using a variety of”
Understanding web browsing behaviors through Weibull analysis of dwell time (2010, background): “, [6, 21, 5, 10]), and revealed that document dwell time (i.”
Information Retrieval Evaluation (2011, methodology): “[70] detailing methods of collecting and then using the implicit user behavior available in web logs.”
Short-Term Satisfaction and Long-Term Coverage: Understanding How Users Tolerate Algorithmic Exploration (2018, background): “There has been substantially more work on understanding implicit feedback signals in information retrieval (IR) [6, 11, 23, 31].”
How does clickthrough data reflect retrieval quality? (2008, background): “For example, there may be sophisticated combinations of various absolute metrics that are more reliable than any single metric [10, 8].”
Denoising Implicit Feedback for Recommendation (2020, background): “[3] utilized multiple implicit users’ behaviors (e.”
A probabilistic method for inferring preferences from clicks (2011, background): “[7] found that combining several implicit indicators can improve accuracy, though it remains well below that of explicit feedback.”
Detecting duplicate web documents using clickthrough data (2011, background): “[19], to obtain cleaner data we consider a result to be clicked on if a user clicked on this result, and did not click on any other result or issue another search query within 30 seconds.”
Efficient multiple-click models in web search (2009, background): “An earlier work in evaluating these implicit measures appeared in [10].”
Understanding User Behavior Through Log Data and Analysis (2014, background): “Logs can also be created with code at the client end, but this requires users to download some sort of logging program and for that program to send data back to the server (Capra, 2011 ; Fox et al., 2005 ).”