Random walks on the click graph

By Nick Craswell, Martin Szummer. Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 2007

518 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/3f53bbebf00bac61b89dfbba4c91505a7d6c783a

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

Citations in the corpus

Learning query intent from regularized click graphs (2008, methodology): “When a click graph is noisy or sparse (as is usually the case [ 8 ]), such regularization would prevent a click graph from propagating erroneous labels.”
Graph-based term weighting for information retrieval (2012, background): “graph theoretic approaches to IR is in the context of social or collaborative networks and recommender systems (Craswell and Szummer 2007; Kleinberg 2006; Konstas et al. 2009; Noh et al. 2009; Schenkel et al. 2008).”
Deep Multimodal Distance Metric Learning Using Click Constraints for Image Ranking (2017, methodology): “For image analysis, Craswell and Szummer [44] proposed a method which adopted clicks for image retrieval, but this approach neglects visual features.”
Learning of Multimodal Representations With Random Walks on the Click Graph (2016, background): “[20] propose a Markov random walk model to a large click log for finding relevant documents, including those that as-yet unclicked for a query, without analyzing the query content or image content.”
Fast personalized PageRank on MapReduce (2011, methodology): “Using random walks on query-click graphs has been shown to be very successful in finding related queries [7].”
Click-through-based cross-view learning for image search (2014, methodology): “For image search, click-through data has been found to be very reliable [6][11].”
Protecting privacy when mining and sharing user data (2012, methodology): “While a query click graph is not as rich a dataset as the actual search log, many computations can still be performed on the click graph with results similar to the computations starting with the actual search log [18, 40].”
Query suggestion using hitting time (2008, methodology): “[9] proposed a ranking function which is basically computing the n-step transition probability from the original vertex to the target.”
Semantic Matching in Search (2014, methodology): “Another state-of-the-art method for finding similar queries from a click-through graph is to perform random walk on the graph, proposed by Craswell and Szummer [50].”
LambdaMerge: merging the results of query reformulations (2011, methodology): “Candidate reformulations for a query can be collected from sources such as user session rewrites [16, 17], anchortext [18, 9], word co-occurrence in documents [27, 30], click graphs [6, 32] or combination of multiple sources [13].”
The query-flow graph: model and applications (2008, background): “Craswell and Szummer [9] describe a Markov random walk model for ranking documents.”
On social networks and collaborative recommendation (2009, methodology): “Craswell and Szummer [4] study a click-through data graph in order to perform item recommendation.”