Filters








72 Hits in 7.2 sec

Diverse retrieval via greedy optimization of expected 1-call@k in a latent subtopic relevance model

Scott Sanner, Shengbo Guo, Thore Graepel, Sadegh Kharazmi, Sarvnaz Karimi
2011 Proceedings of the 20th ACM international conference on Information and knowledge management - CIKM '11  
In this paper, we proceed one step further and show theoretically that greedily optimizing expected 1-call@k w.r.t. a latent subtopic model of binary relevance leads to a diverse retrieval algorithm sharing  ...  As such, the derivation presented here for expected 1-call@k provides a novel theoretical perspective on the emergence of diversity via a latent subtopic model of relevance -an idea underlying both ambiguous  ...  of Excellence program.  ... 
doi:10.1145/2063576.2063869 dblp:conf/cikm/SannerGGKK11 fatcat:aq3mrlkm4jf3fjvsvi7hkbejka

On the mathematical relationship between expected n-call@k and the relevance vs. diversity trade-off

Kar Wai Lim, Scott Sanner, Shengbo Guo
2012 Proceedings of the 35th international ACM SIGIR conference on Research and development in information retrieval - SIGIR '12  
It has been previously noted that optimization of the n-call@k relevance objective (i.e., a set-based objective that is 1 if at least n documents in a set of k are relevant, otherwise 0) encourages more  ...  In this work, we explicitly derive the mathematical relationship between expected n-call@k and the relevance vs. diversity trade-off --- through fortuitous cancellations in the resulting combinatorial  ...  -Need a model for this -In particular, one that models query and document ambiguity (via latent topics) • Since we hypothesize that topic ambiguity underlies the need for diversity Latent subtopic  ... 
doi:10.1145/2348283.2348497 dblp:conf/sigir/LimSG12 fatcat:3ify37yvnncc5o6ye6cnlbcodq

Revisiting the cluster-based paradigm for implicit search result diversification

Hai-Tao Yu, Adam Jatowt, Roi Blanco, Hideo Joho, Joemon M. Jose, Long Chen, Fajie Yuan
2018 Information Processing & Management  
The key idea is that: for a specific query, we expect to maximize the overall relevance of the k exemplar documents.  ...  In particular, how the methods relying on greedy strategies impact the performance of implicit SRD, and how a particular diversification model should be fine-tuned.  ...  Later on, the greedy optimization of Exp-1-call@k [40] for implicit SRD was proposed.  ... 
doi:10.1016/j.ipm.2018.03.003 fatcat:hq5ln4f3kfhrvjr735quvvrnnq

Mining subtopics from different aspects for diversifying search results

Chieh-Jen Wang, Yung-Wei Lin, Ming-Feng Tsai, Hsin-Hsi Chen
2012 Information retrieval (Boston)  
Furthermore, we propose a diversified retrieval model to rank documents with respect to the mined subtopics for balancing relevance and diversity.  ...  Experimental results show that the proposed subtopic-based diversification algorithm significantly outperforms the state-of-the-art models in the TREC09 and TREC10 Web Track diversity tasks.  ...  We also formulated the baseline models as an optimization problem, and the greedy algorithm was used to optimize both the relevance function and the diversity function.  ... 
doi:10.1007/s10791-012-9215-y fatcat:l35ktd43zjdnffmy7enlmxrjse

Multiple Queries as Bandit Arms

Cheng Li, Paul Resnick, Qiaozhu Mei
2016 Proceedings of the 25th ACM International on Conference on Information and Knowledge Management - CIKM '16  
In the presence of rate limits, the active queries take turns accessing the index to retrieve another "page" of results. Turns are assigned by a multi-armed bandit based on user feedback.  ...  Significant improvement is observed both when the subtopic queries are known in advance and when the queries are generated in a user-interactive process.  ...  This work is partially supported by the National Science Foundation under grant numbers IIS-0968489 and IIS-1054199 and by the National Institutes of Health under grant number NLM 2R01LM010681-05.  ... 
doi:10.1145/2983323.2983816 dblp:conf/cikm/LiRM16 fatcat:d4ijehk4q5dizixcgjjytxhxwq

Coverage, redundancy and size-awareness in genre diversity for recommender systems

Saúl Vargas, Linas Baltrunas, Alexandros Karatzoglou, Pablo Castells
2014 Proceedings of the 8th ACM Conference on Recommender systems - RecSys '14  
We also propose an efficient greedy optimization technique to optimize Binomial diversity.  ...  There is increasing awareness in the Recommender Systems field that diversity is a key property that enhances the usefulness of recommendations.  ...  The basis for measuring this proportionality is the so-called disproportionality metric, defined as: DP = s 1 vs≥k R s (vs − k R s ) 2 + 1 2 n 2 NR (5) where vs is the expected number of documents that  ... 
doi:10.1145/2645710.2645743 dblp:conf/recsys/VargasBKC14 fatcat:cvryyozhcngcrmpjmxqo3odgai

Diverse Yet Efficient Retrieval using Locality Sensitive Hashing

Vidyadhar Rao, Prateek Jain, C.V. Jawahar
2016 Proceedings of the 2016 ACM on International Conference on Multimedia Retrieval - ICMR '16  
We extend our method to the problem of multi-label prediction, where the goal is to output a diverse and accurate set of labels for a given document in real-time.  ...  Typical retrieval systems have three requirements: a) Accurate retrieval i.e., the method should have high precision, b) Diverse retrieval, i.e., the obtained set of points should be diverse, c) Retrieval  ...  More specifically, it is theoretically shown [32] that greedily optimizing expected 1call@k w.r.t a latent subtopic model of binary relevance leads to a diverse retrieval algorithm that shares many  ... 
doi:10.1145/2911996.2911998 dblp:conf/mir/RaoJJ16 fatcat:eb2lnpa2ovhtzdole5vp2o5wue

Term level search result diversification

Van Dang, Bruce W. Croft
2013 Proceedings of the 36th international ACM SIGIR conference on Research and development in information retrieval - SIGIR '13  
On the other hand, the explicit approach models the set of query topics, or aspects.  ...  Instead of modeling the set of query aspects, which are typically represented as coherent groups of terms, our approach uses terms without the grouping.  ...  To the best of our knowledge, our method is the first one that can provide statistically significant improvement over standard relevance-based retrieval models in both relevance and diversity measures,  ... 
doi:10.1145/2484028.2484095 dblp:conf/sigir/DangC13 fatcat:lyq7pke7qvee5hepx5bfbuilja

Adapting Markov Decision Process for Search Result Diversification

Long Xia, Jun Xu, Yanyan Lan, Jiafeng Guo, Wei Zeng, Xueqi Cheng
2017 Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval - SIGIR '17  
In this paper, we present a novel diverse ranking model on the basis of continuous state Markov decision process (MDP) in which the user perceived utility is modeled as a part of the MDP state.  ...  . us, to select an optimal document for a position, it is critical for a diverse ranking model to capture the utility of information the user have perceived from the preceding documents.  ...  INTRODUCTION In many information retrieval tasks, one important goal involves providing search results that covers a wide range of topics for a search query, called search result diversi cation [1] .  ... 
doi:10.1145/3077136.3080775 dblp:conf/sigir/XiaXLGZC17 fatcat:abtw3h3jbbfrfazs3iurbj2try

Search Result Diversification in Short Text Streams

Shangsong Liang, Emine Yilmaz, Hong Shen, Maarten De Rijke, W. Bruce Croft
2017 ACM Transactions on Information Systems  
To capture the changes of topics and the probabilities of documents for a given query at a specific time in a short text stream, we propose a dynamic Dirichlet multinomial mixture topic model, called D2M3  ...  Diversifying search results in short text streams is more challenging than in the case of long documents, as it is difficult to capture the latent topics of short documents.  ...  To copy otherwise, to republish, to post on servers, to redistribute to lists, or to use any component of this work in other works requires prior specific permission and/or a fee.  ... 
doi:10.1145/3057282 fatcat:magvfcd3xrgmxh6g35ocu3c4ba

Dynamic Search – Optimizing the Game of Information Seeking [article]

Zhiwen Tang, Grace Hui Yang
2021 arXiv   pre-print
The paper ends with a review of evaluations of dynamic search systems.  ...  To position dynamic search in a larger research landscape, the article discusses in detail its relationship to related research topics and disciplines.  ...  Any opinions, findings, conclusions, or recommendations expressed in this paper are of the authors, and do not necessarily reflect those of the sponsor.  ... 
arXiv:1909.12425v2 fatcat:mdby4xq4jrg4pm6jpxslsv2sri

Enhancing diversity, coverage and balance for summarization through structure learning

Liangda Li, Ke Zhou, Gui-Rong Xue, Hongyuan Zha, Yong Yu
2009 Proceedings of the 18th international conference on World wide web - WWW '09  
In this paper, we consider extract-based summarization emphasizing the following three requirements: 1) diversity in summarization, which seeks to reduce redundancy among sentences in the summary; 2) sufficient  ...  Experiments on the DUC2001 data sets demonstrate significant performance improvements in terms of F 1 and ROUGE metrics.  ...  LR: LR can be regarded as a discriminative version of NB. It is employed to model the posterior probabilities of k classes via linear functions in output variables.  ... 
doi:10.1145/1526709.1526720 dblp:conf/www/LiZXZY09 fatcat:ndqpm7xcbzhb5a6btmo4zyfxbi

Ranking ideas for diversity and quality [article]

Faez Ahmed, Mark Fuge
2017 arXiv   pre-print
These rank metrics open up the use of multi-objective optimization to describe trade-offs between diversity and quality in ranked lists.  ...  incurs only a marginal increase in the number of designs considered.  ...  [40] for multidocument summarization, which rewards diversity of a set of items as shown below: Div 1 (S) = K k=1 j∈S∩P k 1 N × M i∈P k L i,j (1) Here, V = v 1 , ..., v n is the set of all N items in  ... 
arXiv:1709.02063v1 fatcat:f2dgx52sxvgh5b6hanzwzoieci

Result diversification based on query-specific cluster ranking

Jiyin He, Edgar Meij, Maarten de Rijke
2011 Journal of the American Society for Information Science and Technology  
a high percentage of relevant documents.  ...  Most relevant documents should be contained in a small number of high-quality clusters, while there should be no dominantly large clusters.  ...  They show that in such a scenario, it is more effective to optimize the expected value of a given metric and to rank documents in such a way that the probability of finding at least a relevant document  ... 
doi:10.1002/asi.21468 fatcat:rapy4pagqvcfpio725ek4o5qfi

Exploring topic structure

Jiyin He
2012 SIGIR Forum  
Exploring topic structure: Coherence, diversity and relatedness General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or  ...  Exploring topic structure: Coherence, diversity and relatedness He, J. Link to publication Citation for published version (APA): He, J. (2011).  ...  We write D n q for the top n documents retrieved in response to q. In D n q we identify a set of K clusters, C = {c k } K k=1 .  ... 
doi:10.1145/2215676.2215690 fatcat:mh34cmstt5cr3phgj2cltgikie
« Previous Showing results 1 — 15 out of 72 results