36,237 Hits in 7.5 sec

On the suitability of diversity metrics for learning-to-rank for diversity

Rodrygo L.T. Santos, Craig Macdonald, Iadh Ounis
2011 Proceedings of the 34th international ACM SIGIR conference on Research and development in Information - SIGIR '11  
Although evaluation metrics that reward coverage and penalise redundancy provide intuitive objective functions for learning a diverse ranking, it is unclear whether they are the most effective.  ...  In this paper, we contrast the suitability of relevance and diversity metrics as objective functions for learning a diverse ranking.  ...  While ERR and nDCG are established metrics for web search evaluation, ERR-IA and α-nDCG are the primary evaluation metrics in the diversity task of the TREC 2010 Web track.  ... 
doi:10.1145/2009916.2010111 dblp:conf/sigir/SantosMO11a fatcat:drxbbtuogvbjha7hlzekeoafum

Redundancy, diversity and interdependent document relevance

Filip Radlinski, Paul N. Bennett, Ben Carterette, Thorsten Joachims
2009 SIGIR Forum  
methods, and evaluation techniques.  ...  The goal of the Redundancy, Diversity, and Interdependent Document Relevance workshop was to explore how ranking, performance assessment and learning to rank can move beyond the assumption that the relevance  ...  We are particularly thankful for the feedback received from the workshop chair (Diane Kelly) and general chairs (James Allan and Javed Aslam), and their quick responses to questions.  ... 
doi:10.1145/1670564.1670572 fatcat:oip27uzovrbl3cw3wr4sxx2qhy

Search Result Diversification

Rodrygo L. T. Santos, Craig Macdonald, Iadh Ounis
2015 Foundations and Trends in Information Retrieval  
In addition, we introduced different evaluation frameworks and the most used evaluation metrics based on them.  ...  Diversity-oriented ranking has been proposed as a means to overcome ambiguity and redundancy during the search process.  ... 
doi:10.1561/1500000040 fatcat:g4vxxda6u5aaljeqg6n47lushu

Explicit relevance models in intent-oriented information retrieval diversification

Saúl Vargas, Pablo Castells, David Vallet
2012 Proceedings of the 35th international ACM SIGIR conference on Research and development in information retrieval - SIGIR '12  
We illustrate this by modeling tolerance to redundancy as an explicit configurable parameter, which can be set to better suit the characteristics of the IR task, or the evaluation metrics, as we illustrate  ...  We report experiments on search and recommendation tasks showing competitive or better performance than the original diversification algorithms.  ...  Diversity-enhancement methods have been developed [1, 7, 21, 27, 29] , diversity evaluation methodologies and metrics have been proposed [1, 8, 24, 29] , and a diversity task has been included in the  ... 
doi:10.1145/2348283.2348297 dblp:conf/sigir/VargasCV12 fatcat:tcs3vs7o3jbabkfffr24tkf7wa

Information filtering based on corrected redundancy-eliminating mass diffusion

Xuzhen Zhu, Yujie Yang, Guilin Chen, Matus Medo, Hui Tian, Shi-Min Cai, Jian-Guo Liu
2017 PLoS ONE  
The used similarity metrics often suffer from similarity redundancies arising from correlations between objects' attributes.  ...  Based on an unweighted undirected object-user bipartite network, we propose a Corrected Redundancy-Eliminating similarity index (CRE) which is based on a spreading process on the network.  ...  Acknowledgments This work was supported by National Natural Science Foundation of China (Nos. 61602048, 61673086, and 61471060), the support project of Sichuan Provincial Science and Technology Department  ... 
doi:10.1371/journal.pone.0181402 pmid:28749976 pmcid:PMC5531469 fatcat:3denziffcfaafabrtmaqts2j2e

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  
In this work we propose a new Binomial framework for defining genre diversity in recommender systems that takes into account three key properties: genre coverage, genre redundancy and recommendation list  ...  We show that methods previously proposed for measuring and enhancing recommendation diversity -including those adapted from search result diversification-fail to address adequately these three properties  ...  We evaluate the diversity of the recommendations with respect to genres by our proposed Binomial Diversity metric (BinomDiv), plus the diversity metrics presented in Section 2.2: EILD, ERR-IA and CPR.  ... 
doi:10.1145/2645710.2645743 dblp:conf/recsys/VargasBKC14 fatcat:cvryyozhcngcrmpjmxqo3odgai

Topic Pages: An Alternative to the Ten Blue Links

Niranjan Balasubramanian, Silviu Cucerzan
2010 2010 IEEE Fourth International Conference on Semantic Computing  
Automatic and manual evaluations on biographical topics show that topic pages built by our system compare favorably to regular Web search results and to MDS-style summaries of the Web results on all metrics  ...  Topic pages explicitly aggregate information across documents, filter redundancy, and promote diversity of topical aspects.  ...  However, because search snippets are short and generated independently of each other, they provide only superficial and possibly redundant coverage of topical aspects.  ... 
doi:10.1109/icsc.2010.96 dblp:conf/semco/BalasubramanianC10 fatcat:kxrayzdlhfdpxn3okmnsl4or44

The use of MMR, diversity-based reranking for reordering documents and producing summaries

Jaime Carbonell, Jade Goldstein
1998 Proceedings of the 21st annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '98  
Preliminary results indicate some bene ts for MMR diversity ranking in document retrieval and in single document summarization.  ...  The Maximal Marginal Relevance (MMR) criterion strives to reduce redundancy while maintaining query relevance in re-ranking retrieved documents and in selecting appropriate passages for text summarization  ...  Concluding Remarks We have shown that MMR ranking provides a useful and bene cial manner of providing information to the user by allowing the user to minimize redundancy.  ... 
doi:10.1145/290941.291025 dblp:conf/sigir/CarbonellG98 fatcat:eer6v4scqnhk7pj3rxu45me4cq

Independent Verification and Diversity: The Echelons for Assurance of Cyber Physical Systems Safety

Vyacheslav Kharchenko
2020 International Workshop on Information-Communication Technologies & Embedded Systems  
Independent verification and validation (IV&V) and D3 (Defence-in-Depth and Diversity) approach are two echelons for protection of CPSs against cyber and physical attacks and failures caused by physical  ...  Application of diversity for safety and security assurance is discussed.  ...  Metrics of relative number of MCCFs and DCCFs (see Figure 2 Application of defence-in-depth and diversity for safety and security assurance Classification of different diversity types and D3 in general  ... 
dblp:conf/ictes/Kharchenko20 fatcat:barzdvrbivdolff77hyetl4qp4

Modelling efficient novelty-based search result diversification in metric spaces

Veronica Gil-Costa, Rodrygo L.T. Santos, Craig Macdonald, Iadh Ounis
2013 Journal of Discrete Algorithms  
We investigate three different approaches: pivoting-based, clustering-based, and permutation-based.  ...  Thorough experiments using two TREC test collections for diversity evaluation, as well as a large sample of the query stream of a commercial search engine show that our approaches perform at least as effectively  ...  To evaluate our approach in different metric spaces, we experiment with two standard test collections for diversity evaluation, comprising both Web and newswire documents.  ... 
doi:10.1016/j.jda.2012.07.004 fatcat:mfhv2qqeqvepblydlukmg2o6mq

Sequence Clustering Methods and Completeness of Biological Database Search

Qingyu Chen, Xiuzhen Zhang, Yu Wan, Justin Zobel, Karin Verspoor
2017 International Joint Conference on Artificial Intelligence  
Users then search against the resulting non-redundant database, which is typically comprised of one representative sequence per cluster, and expand search results by exploring records from matching clusters  ...  Compared to direct search of original databases, the search results are expected to be more diverse are also more complete.  ...  Acknowledgments We appreciate the advice of the NCBI BLAST team on BLAST related commands and parameters.  ... 
dblp:conf/ijcai/0001ZWZV17 fatcat:j4cdmavj3rbhpfjyp4jn3yerfi

An automated framework for characterizing and subsetting GPGPU workloads

Vignesh Adhinarayanan, Wu-chun Feng
2016 2016 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS)  
In this study, we use our framework to identify redundancy in the recently released SPEC ACCEL OpenCL benchmark suite using a few architecture-dependent metrics.  ...  Sometimes researchers spend unnecessary time evaluating redundant workloads, which is particularly problematic for timeconsuming studies involving simulation.  ...  ACKNOWLEDGMENT This work was supported in part by NSF I/UCRC IIP-0804155 and IIP-1266245 via the NSF Center for High-Performance Reconfigurable Computing.  ... 
doi:10.1109/ispass.2016.7482105 dblp:conf/ispass/AdhinarayananF16 fatcat:uxwgfxtgkjcndkfr4hbwps757i

Towards Unified Metrics for Accuracy and Diversity for Recommender Systems

Javier Parapar, Filip Radlinski
2021 Fifteenth ACM Conference on Recommender Systems  
The proposed metric combines topical diversity and accuracy, and we show it to satisfy a set of desired properties that we formulate axiomatically.  ...  We propose a novel adaptation of a unified metric, derived from one commonly used for search system evaluation, to Recommender Systems.  ...  ACKNOWLEDGMENTS The first author wants to thank Google's Visiting Researcher Program and Google AI London team for their warm welcome.  ... 
doi:10.1145/3460231.3474234 fatcat:kzcdx4elkbchrg4fufs7uefdey

A Meta-Evaluation of Evaluation Methods for Diversified Search [chapter]

Suneel Kumar Kingrani, Mark Levene, Dell Zhang
2018 Lecture Notes in Computer Science  
Prior to making use of such evaluation methods, it is important to have a good understanding of how diversity and relevance contribute to the performance metric of each method.  ...  In this paper, we use the statistical technique ANOVA to analyse and compare three representative evaluation methods for diversified search, namely α-nDCG, MAP-IA, and ERR-IA, on the TREC-2009 Web track  ...  IR metrics [15] to capture both the diversity and the relevance of search results.  ... 
doi:10.1007/978-3-319-76941-7_43 fatcat:bvdbf3s56jedfekpmzqefr6aqm

Improving web search results using affinity graph

Benyu Zhang, Hua Li, Yi Liu, Lei Ji, Wensi Xi, Weiguo Fan, Zheng Chen, Wei-Ying Ma
2005 Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '05  
Specifically, the algorithm achieves 31% improvement in diversity and 12% improvement in information richness relatively within the top 10 search results.  ...  In this paper, we propose a novel ranking scheme named Affinity Ranking (AR) to re-rank search results by optimizing two metrics: (1) diversity --which indicates the variance of topics in a group of documents  ...  Four researchers in web search and mining area are hired to independently evaluate the experimental results.  ... 
doi:10.1145/1076034.1076120 dblp:conf/sigir/ZhangLLJXFCM05 fatcat:eh4xkldz6nhi3mgckx2mvvessu
« Previous Showing results 1 — 15 out of 36,237 results