Improving web search ranking by incorporating user behavior information

By Eugene Agichtein, Eric Brill, Susan T. Dumais. Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 2006

482 citations recorded by the source corpus.

Subfields: Information Systems

View Paper: https://www.semanticscholar.org/paper/3dba70c6d08281a6cc6f4fe4647a6d0db779f4eb

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

Citations in the corpus

Cyberchondria: Studies of the escalation of medical concerns in Web search (2009, methodology): “For example, ranking methods employed by search services may take user clicks and dwells on Web pages as an indication that the page is relevant to the adjacent query [Agichtein et al. 2006].”
Beyond position bias: examining result attractiveness as a source of presentation bias in clickthrough data (2010, background): “One can also use implicit feedback to design new signals (or features) to be used by ranking functions [ 1 , 5].”
SLiMFast: Guaranteed Results for Data Fusion and Source Reliability (2015, background): “lications from different elds, such as that of information retrieval, have traditionally used external signals , i.e., descriptive statistics or metadata of the sources, to characterize their quality [3, 4]. For example, trafc data of a webpage are commonly used to assess its credibility, while appl”
Learning diverse rankings with multi-armed bandits (2008, background): “With only few exceptions (e.g. Radlinski & Joachims, 2007), previous work does not consider such an online approach.”
Methods for Evaluating Interactive Information Retrieval Systems with Users (2009, background): “[3], Bilenko and White [28], and Jansen and Spink [146].”
Learning to Rank for Information Retrieval and Natural Language Processing (2011, unknown): “Web pages usually contain a number of fields (metadata streams) such as title, anchor texts, URL, extracted title [50, 51], and associated queries in click-through data [3].”
Revolt: Collaborative Crowdsourcing for Labeling Machine Learning Datasets (2017, background): “…experts for manual annotation [51], extracting relations from readily available sources (e.g., identifying bodies of text in parallel online translations [46, 13]), and automatically generating labels based on user behaviors (e.g., using dwell time to implicitly mark search result relevance [2]).”
Learning to rank for information retrieval (2009, unknown): “In our experiments, the query-document feature values were normalized on a per query basis to the [0, 1] interval using the linear scaling suggested by the producers of the LETOR dataset and no additional feature selection or processing was done.”
Evaluation of contextual information retrieval effectiveness: overview of issues and research (2010, methodology): “It refers to the user’s levels of expertise [73 ]a nduser interests either short-term ones [ 1 ,43,57,91] or long-term ones [70,93,103].”
Evaluating the accuracy of implicit feedback from clicks and query reformulations in Web search (2007, background): “…statistics can be more accurate than preferences from individual clicks, and they demonstrated how clicks in addition to other implicit feedback signals (e.g., dwell time) can provide signi.cant improvements of retrieval accuracy in real-world search engines [Agichtein et al. 2006a].”