Filters








10,347 Hits in 3.6 sec

Random walks on the click graph

Nick Craswell, Martin Szummer
2007 Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '07  
We conduct experiments on click logs from image search, comparing our ('backward') random walk model to a different ('forward') random walk, varying parameters such as walk length and self-transition probability  ...  We apply a Markov random walk model to a large click log, producing a probabilistic ranking of documents for a given query.  ...  Ours is a query-dependent backward random walk on the click graph, where walk length is moderated by t and s.  ... 
doi:10.1145/1277741.1277784 dblp:conf/sigir/CraswellS07 fatcat:vvhav7ntqzhnvmtfho7phr3dpe

Dr. Searcher and Mr. Browser

Barbara Poblete, Carlos Castillo, Aristides Gionis
2008 Proceeding of the 17th ACM conference on Information and knowledge mining - CIKM '08  
Our most important motivation is to model in a unified way the two main activities of users on the Web: searching and browsing, and at the same time to analyze the effects of random walks on this new graph  ...  In particular stationary distribution scores derived from the random walks on the combined graph can be used as an indicator of whether structural or usage data are more reliable in different situations  ...  The authors thank Claudio Corsi from the University of Pisa for his help with data preprocessing. Also we thank Debora Donato and Vanessa Murdock from Yahoo!  ... 
doi:10.1145/1458082.1458231 dblp:conf/cikm/PobleteCG08 fatcat:yblizjicofdnnllkpmz55vbmqe

Exploiting click-through data for entity retrieval

Bodo Billerbeck, Gianluca Demartini, Claudiu Firan, Tereza Iofciu, Ralf Krestel
2010 Proceeding of the 33rd international ACM SIGIR conference on Research and development in information retrieval - SIGIR '10  
We compare results using click graphs and session graphs and present an evaluation test set making use of Wikipedia "List of" pages.  ...  We present an approach for answering Entity Retrieval queries using click-through information in query log data from a commercial Web search engine.  ...  We apply the results of query log analysis to ER by performing random walks on click and session graphs.  ... 
doi:10.1145/1835449.1835624 dblp:conf/sigir/BillerbeckDFIK10 fatcat:nqurnurrnnfazpqhbitw7cdpkm

Ranking Entities Using Web Search Query Logs [chapter]

Bodo Billerbeck, Gianluca Demartini, Claudiu S. Firan, Tereza Iofciu, Ralf Krestel
2010 Lecture Notes in Computer Science  
We use Markov random walks on (1) Click Graphs -built from clickthrough data -and on (2) Session Graphs -built from user session information.  ...  Searching for entities is an emerging task in Information Retrieval for which the goal is finding well defined entities instead of documents matching the query terms.  ...  Walking the Graphs for Entity Ranking Similarly to [5] we perform a Markov random walk on the click and session graphs in order to find relevant results for query q.  ... 
doi:10.1007/978-3-642-15464-5_28 fatcat:awqjas7azrcovanarwwlim36zq

Behavioral classification on the click graph

Martin Szummer, Nick Craswell
2008 Proceeding of the 17th international conference on World Wide Web - WWW '08  
We show how to perform classification using random walks on this graph, and two methods for estimating classifier parameters.  ...  We choose a click graph sampled from two weeks of image search activity, and the task of "adult" filtering: identifying content in the graph that is inappropriate for minors.  ...  The walk captures the transitivity of class similarity on the graph: if A is co-clicked with B and B is co-clicked with C, then A is also likely to be related to C.  ... 
doi:10.1145/1367497.1367746 dblp:conf/www/SzummerC08 fatcat:ihrla7zlwfhonmv7gokcnp5pwe

Optimal rare query suggestion with implicit user feedback

Yang Song, Li-wei He
2010 Proceedings of the 19th international conference on World wide web - WWW '10  
Hence, our framework optimally combines both the click and skip information from users and uses a random walk model to optimize the query correlation.  ...  Our model specifically optimizes two parameters: (1) the restarting (jumping) rate of random walk, and (2) the combination ratio of click and skip information.  ...  Query audi parts and audi bodywork are not correlated if only performs random walk on the click graph, but will be highly correlation if random walk is performed on the skip graph.  ... 
doi:10.1145/1772690.1772782 dblp:conf/www/SongH10 fatcat:b2m6gsugdfajddkzdaialokb2m

Enriched Network Embeddings for News Recommendation

Janu Verma
2019 ACM Conference on Recommender Systems  
The random walk based graph embeddings are used to learn latent representation for users, articles and named entities in the same space.  ...  We propose a recommendation system based on the binary classification problem which takes as input a combination of the user, item and entity embeddings and computes the probability of the user clicking  ...  The random walks on the bipartite graph have paths of the form 𝑈 𝑠𝑒𝑟 → 𝐴𝑟𝑡𝑖𝑐𝑙𝑒 → 𝑈 𝑠𝑒𝑟 The random walks thus generated are sequences of the nodes, which can be thought as 'sentences' in  ... 
dblp:conf/recsys/Verma19 fatcat:adquxd6a4vhitoovtpj2nnq42i

Click-boosting random walk for image search reranking

Xiaopeng Yang, Yongdong Zhang, Ting Yao, Zheng-Jun Zha, Chong-Wah Ngo
2013 Proceedings of the Fifth International Conference on Internet Multimedia Computing and Service - ICIMCS '13  
A new algorithm, named Clickboosting Random Walk, is proposed. The algorithm utilizes clicked images to locate similar images that are not clicked, and reranks them by random walk.  ...  This paper explores the use of click-through data, which can be viewed as the footprints of user searching behavior, as an effective means of understanding query, for providing the basis on identifying  ...  Second, it performs random walk on an image graph based on the click-boosted ranked results, where nodes in the graph represent images and edges between them are weighted by visual similarities.  ... 
doi:10.1145/2499788.2499810 dblp:conf/icimcs/YangZYZN13 fatcat:4khgavq4vfgf5mmmzc63omotji

Discovering missing click-through query language information for web search

Xing Yi, James Allan
2011 Proceedings of the 20th ACM international conference on Information and knowledge management - CIKM '11  
We further propose combining this content based approach with the random walk approach on the click graph to further reduce click-through sparseness for search.  ...  The combination approach that uses discovered click-through features from both random walk and the content based approach can further improve search performance.  ...  Finding More Click-associated Queries by Random Walk on Click Graph We start by reviewing the random walk approach that uses co-clicks in the click-through data to discover plausible missing clicks for  ... 
doi:10.1145/2063576.2063604 dblp:conf/cikm/YiA11 fatcat:or4ulroflnexjl6hpgxmoepeyy

Recommending High Utility Query via Session-Flow Graph [chapter]

Xiaofei Zhu, Jiafeng Guo, Xueqi Cheng, Yanyan Lan, Wolfgang Nejdl
2013 Lecture Notes in Computer Science  
In this paper, we propose a novel utility query recommendation approach based on absorbing random walk on the session-flow graph, which can learn queries' utility by simultaneously modeling both users'  ...  reformulation behaviors and click behaviors.  ...  ., run an random walk on the click-through graph [7] .  ... 
doi:10.1007/978-3-642-36973-5_54 fatcat:xn2u4wbwwfgszn5bfnyqdcsyga

Organizing User Search Histories

Heasoo Hwang, H. W. Lauw, L. Getoor, A. Ntoulas
2012 IEEE Transactions on Knowledge and Data Engineering  
To better support users in their long-term information quests on the web, search engines keep track of their queries and clicks while searching online.  ...  Users are increasingly pursuing complex task-oriented goals on the web, such as making travel arrangements, managing finances, or planning purchases.  ...  In Craswell and Szummer [37] , a Markov random walk was applied on the click graph to improve ranking. In Fuxman et al.  ... 
doi:10.1109/tkde.2010.251 fatcat:d67ympyz6zhvndfljq5kdf3mpe

Personalized Query Results using User Search Logs
English

Santhi Kolli, V.Rama chandran, Ru pa
2013 International Journal of Engineering Trends and Technoloy  
Random walk propagation over the query fusion graph methods support complex search quests in IR systems at reduced times.  ...  We propose to implement Random walk propagation methods that can construct user profiles based on the credentials obtained from their prior search history repositories.  ...  Uses random walk propagation over the query fusion graph instead of time-based and keyword similarity based approaches.  ... 
doi:10.14445/22315381/ijett-v4i9p197 fatcat:t6pv2fa3szgkdkamh6voiquvna

Recommendation Framework Combining User Interests with Fashion Trends in Apparel Online Shopping

Ok, Lee, Kim
2019 Applied Sciences  
The ratings are combined with a network constructed by an item click trend, which serves as a personalized recommendation through a random walk.  ...  An empirical evaluation on a large-scale real-world dataset obtained from an apparel retailer demonstrates the effectiveness of our method.  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/app9132634 fatcat:sub7y2joufcf7ew2bf2rxbhysi

Using the wisdom of the crowds for keyword generation

Ariel Fuxman, Panayiotis Tsaparas, Kannan Achan, Rakesh Agrawal
2008 Proceeding of the 17th international conference on World Wide Web - WWW '08  
We formulate the problem as a semi-supervised learning problem, and propose algorithms within the Markov Random Field model.  ...  We identify queries related to a campaign by exploiting the associations between queries and URLs as they are captured by the user's clicks.  ...  Acknowledgements: We would like to thank Marc Najork for providing the tools that we used for storing and processing the click logs, and Alan Halverson for his help in using these tools.  ... 
doi:10.1145/1367497.1367506 dblp:conf/www/FuxmanTAA08 fatcat:25drxkjna5fypcq6s5wcx7iuju

Towards Exploiting Implicit Human Feedback for Improving RDF2vec Embeddings [article]

Ahmad Al Taweel, Heiko Paulheim
2020 arXiv   pre-print
It first creates sequences of nodes by performing random walks on the graph. In a second step, those sequences are processed by the word2vec algorithm for creating the actual embeddings.  ...  We show that in some scenarios, RDF2vec utilizing those transition probabilities can outperform both RDF2vec based on random walks as well as the usage of graph internal edge weights.  ...  Since word2vec operates on (word) sequences, several approaches have been proposed which first turn a graph into sequences by performing random walks, and then applying the idea of word2vec to those sequences  ... 
arXiv:2004.04423v1 fatcat:lrcihhaa6fac7nidltmfnn6koa
« Previous Showing results 1 — 15 out of 10,347 results