Click Models for Web Search

By Aleksandr Chuklin, Ilya Markov, Maarten de Rijke. 2015

429 citations recorded by the source corpus.

Abstract With the rapid growth of web search in recent years the problem of modeling its users has started to attract more and more attention of the information retrieval community. This has several motivations. By building a model of user behavior we are essentially developing a better understanding of a user, which ultimately helps us to deliver a better search experience. A model of user behavior can also be used as a predictive device for non-observed items such as document relevance, which makes it useful for improving search result ranking. Finally, in many situations experimenting with real users is just infeasible and hence user simulations based on accurate models play an essential role in understanding the implications of algorithmic changes to search engine results or presentation changes to the search engine result page. In this survey we summarize advances in modeling user click behavior on a web search engine result page. We present simple click models as well as more complex models aimed at...

Subfields: Computer Vision and Pattern Recognition, 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 Computer Vision and Pattern Recognition

In Information Systems

In Information Retrieval & Search

Citations in the corpus

End-to-End Neural Ad-hoc Ranking with Kernel Pooling (2017, methodology): “d scores a good approximation of expert labels, and Testing-DIFF is expected to produce evaluation results similar to expert labels. Testing-RAWisthesimplestclickmodel. Followingthecascade assumption [3], we treat the clicked document in each single-click session as a relevant document, and test whe”
Time-Aware Click Model (2016, methodology): “We first introduce some basic click models for Web search [Chuklin et al. 2015] to show the essential ideas and assumptions of the click model and then introduce some existing click models that can partially handle temporal information.”
Entity-Duet Neural Ranking: Understanding the Role of Knowledge Graph Semantics in Neural Information Retrieval (2018, methodology): “Prior research leverages clicks to model user behaviors and infer reliable relevance signals using click models (Chuklin et al., 2015).”
Unbiased Learning-to-Rank with Biased Feedback (2016, methodology): “Other click models ([5, 3, 1], also see [4]) have been proposed, and are trained to maximize log-likelihood of observed clicks.”
A General Framework for Counterfactual Learning-to-Rank (2018, background): “Other click models ([3, 7, 9], also see [8]) train to maximize loglikelihood of observed clicks, where relevance is modeled as a latent variable that is inferred over multiple instances of the same query.”
MergeDTS (2018, methodology): “[24] and simulate clicks using three configurations of a click model [12]: namely perfect, navigational and informational.”
Bandit Algorithms (2020, methodology): “For more information on click models we recommend the survey paper by Chuklin et al. [2015] and article by Craswell et al. [2008]. Cascading bandits were first studied by Kveton et al. [2015a], who proposed algorithms based on UCB and KL-UCB and prove finitetime instance-dependence upper bounds and ”
Social Collaborative Viewpoint Regression with Explainable Recommendations (2017, background): “As to future work, we plan to explore whether ranking-based strategies that integrate our sCVR model can enhance the performance of item recommendation and whether a combination with interaction data [9] can help to improve performance even further.”
Learning to Rank with Selection Bias in Personal Search (2016, background): “[10] provides a good overview of the latest advances in the field.”
Neural information retrieval: at the end of the early years (2017, background): “…by learning from user behavior, either by introducing, e.g., click-based rankers (Radlinski and Joachims 2005) or, more abstractly, by using models that capture behavioral notions such as examination probability and attractiveness of search results through click models (Chuklin et al. 2015).”
Position Bias Estimation for Unbiased Learning to Rank in Personal Search (2018, methodology): “Though simple, this model has been shown to be as effective as more sophisticated click models [9].”
Convolutional Neural Networks for Soft-Matching N-Grams in Ad-hoc Search (2018, neutral): “Training Labels: e training labels for the Sogou and Bing logs were generated by the DCTR click model from user clicks in the training sessions [5], and training preference pairs were constructed accordingly.”
Equity of Attention: Amortizing Individual Fairness in Rankings (2018, methodology): “These observations have led to a variety of click models (see [8] for a survey), and several methods for biasaware re-ranking (e.”