Filters








3,615 Hits in 3.1 sec

Towards A Fairer Landmark Recognition Dataset [article]

Zu Kim, André Araujo, Bingyi Cao, Cam Askew, Jack Sim, Mike Green, N'Mah Fodiatu Yilla, Tobias Weyand
2021 arXiv   pre-print
We present a stratification approach and analysis which leads to a much fairer coverage of the world, compared to existing datasets.  ...  These relevances are estimated by combining anonymized Google Maps user contribution statistics with the contributors' demographic information.  ...  These estimated relevance scores allow us to create a fairer test dataset, which is used to evaluate models as part of the 2021 edition of the Google Landmark Recognition 2 and Retrieval 3 challenges.  ... 
arXiv:2108.08874v1 fatcat:cj6xpwkm3zao5fugh46dxb25ey

Algorithmic Bias

Colin Wilkie, Leif Azzopardi
2017 Proceedings of the 2017 ACM on Conference on Information and Knowledge Management - CIKM '17  
Algorithmic bias presents a difficult challenge within Information Retrieval.  ...  The analysis finds that the systems that perform better, tend to have a higher chance of retrieving a relevant document rather than a non-relevant document for a topic prior to retrieval, indicating that  ...  In this work, a related hypothesis is proposed; that better performing systems actually exhibit a bias towards the relevant documents for a query, prior to retrieval.  ... 
doi:10.1145/3132847.3133135 dblp:conf/cikm/WilkieA17 fatcat:k7fpsavcaja33otvzokbklwyme

Best and Fairest: An Empirical Analysis of Retrieval System Bias [chapter]

Colin Wilkie, Leif Azzopardi
2014 Lecture Notes in Computer Science  
This is largely motivated by the recent proposal of a new suite of retrieval models based on the Divergence From Independence (DFI) framework.  ...  Here, we consider bias as the extent to which a retrieval model unduly favours certain documents over others because of characteristics within and about the document.  ...  Summary and Conclusions In this paper, we have measured the retrieval bias of a spectrum of retrieval model/weightings to determine which model is the fairest.  ... 
doi:10.1007/978-3-319-06028-6_2 fatcat:w3uigrxom5c4vnsaepdgnpbli4

Training and fairer payments would increase caries prevention in practice

Susan J Carson, Ruth Freeman
2015 Evidence-Based Dentistry  
Factors that drive dentists towards or away from dental caries preventive measures: systematic review and metasummary. PLoS One 2014; 9: e107831.  ...  The review finds that dental education and training is an important factor in influencing dentists' drive towards preventative treatments for caries, with remuneration systems also suggested as having  ...  Conclusions Despite the questionable quality of the included reports the evidence that emerged seems to indicate that further education and training coupled with a fairer pay scheme would be a reasonable  ... 
doi:10.1038/sj.ebd.6401072 pmid:25909928 fatcat:sqz23jieengulnscjixg27x7mu

Knowledge management: An information science perspective

Gashaw Kebede
2010 International Journal of Information Management  
Knowledge management (KM) is an emerging field of specialization in a number of professions, including Information Science (IS).  ...  The paper also aims at contributing towards achieving a consensus among IS professionals on conceptualization, goals, and scope of KM in IS.  ...  For Blair (2002) IM is a component of KM: "Although Knowledge Management is not the same as data or information management, data and information retrieval can be important components of it."  ... 
doi:10.1016/j.ijinfomgt.2010.02.004 fatcat:ejldinp5tfca7njppz3awsycfe

Academic Publishing and its Digital Binds: Beyond the Paywall towards Ethical Executions of Code

Teresa Swist, Liam Magee
2018 Culture Unbound: Journal of Current Cultural Research  
systems which offer more diverse publishing pathways; and, disrupting systemic processes and profits towards more inclusive and equitable conditions.  ...  Together, these mechanisms show how mutually beneficial boundaries can be drawn for designing otherwise: by blocking dominant systems and bargaining for fairer practices; exploring sanctioned and unsanctioned  ...  There is a possibility that Sci-Hub users-especially those not using privacy-enhancing services such as Tor-could have their usage history unmasked and face consequences, both legal or reputational in  ... 
doi:10.3384/cu.2000.1525.1793240 fatcat:e4eijro2djgglkuecw6aaagjru

Who is the fairest of them all? Public attitudes and expectations regarding automated decision-making

Natali Helberger, Theo Araujo, Claes H. de Vreese
2020 Computer Law and Security Review  
Acknowledgment This research was supported by the Research Priority Areas Information & Communication in the Data Society ( https:// www.uva-icds.net/ ), and Communication and its Digital Communication  ...  In law, for example, it can assist in information retrieval, such as the discovery of relevant case law, or the processing of large amounts of legal data, the automated generation of legal texts, the offering  ...  A.2.1. Who is fairer?  ... 
doi:10.1016/j.clsr.2020.105456 fatcat:cnzqftb7nvecxa5td67bzhanxm

Efficiently Estimating Retrievability Bias [chapter]

Colin Wilkie, Leif Azzopardi
2014 Lecture Notes in Computer Science  
Retrievability is the measure of how easily a document can be retrieved using a particular retrieval system.  ...  The extent to which a retrieval system favours certain documents over others (as expressed by their retrievability scores) determines the level of bias the system imposes on a collection.  ...  A wide range of applications exist including improving recall in retrieval systems [5, 6] , improving the effectiveness of pseudo-relevance feedback [4] , detecting bias towards particular organisations  ... 
doi:10.1007/978-3-319-06028-6_82 fatcat:wk45whrvuranha5pksgzmhle6e

Information management education: towards a holistic perspective

Felicite A. Fairer-Wessels
2014 South African Journal of Libraries and Information Science  
Furthermore the current schools of thought in the field of information are mentioned as well as the paradigm underlying information management. The movement towards a new paradigm is debated.  ...  The emergence of information management as a field of study is discussed as largely a response to the information age and the enormous amount of information which has resulted from it.  ...  (Fairer-Wessels 1995) .  ... 
doi:10.7553/65-2-1479 fatcat:hekhgo4lnre2rgek7gig7eksui

A Comparison of Three Major Academic Rankings for World Universities: From a Research Evaluation Perspective

Mu-hsuan Huang
2011 Journal of Library and Information Studies  
This paper introduces three current major university ranking systems.  ...  This paper compares the 2009 ranking results from the three ranking systems.  ...  That is, in terms of research performance, the HEEACT ranking may be a fairer and a more informative ranking system for the majority of the world's universities.  ... 
doi:10.6182/jlis.2011.9(1).001 doaj:82a519528402442ab6527595492e3e51 fatcat:n3p3t4wbvjbwbeedrr66siyqua

Fair Multi-Stakeholder News Recommender System with Hypergraph ranking [article]

Alireza Gharahighehi, Celine Vens, Konstantinos Pliakos
2021 arXiv   pre-print
The results show that the proposed approach counters popularity bias and produces fairer recommendations with respect to authors in two news datasets, at a low cost in precision.  ...  Recommender systems are typically designed to fulfill end user needs. However, in some domains the users are not the only stakeholders in the system.  ...  To assess the accuracy of the model predictions we use precision, which is a standard information retrieval metric.  ... 
arXiv:2012.00387v2 fatcat:zroyahcjevgank2wb2o6ecsup4

Grep-BiasIR: A Dataset for Investigating Gender Representation-Bias in Information Retrieval Results [article]

Klara Krieg and Emilia Parada-Cabaleiro and Gertraud Medicus and Oleg Lesota and Markus Schedl and Navid Rekabsaz
2022 arXiv   pre-print
To facilitate the studies of gender bias in the retrieval results of IR systems, we introduce Gender Representation-Bias for Information Retrieval (Grep-BiasIR), a novel thoroughly-audited dataset consisting  ...  The provided contents by information retrieval (IR) systems can reflect the existing societal biases and stereotypes.  ...  To study this kind of gender bias in retrieval results, we introduce the Gender Representation-Bias for Information Retrieval (Grep-BiasIR) dataset.  ... 
arXiv:2201.07754v2 fatcat:l4hy7dyv3jfg3pgjjxxudi4aku

Are Gender-Neutral Queries Really Gender-Neutral? Mitigating Gender Bias in Image Search [article]

Jialu Wang and Yang Liu and Xin Eric Wang
2021 arXiv   pre-print
on mutual information to debias multimodal representations of pre-trained models.  ...  We study a unique gender bias in image search in this work: the search images are often gender-imbalanced for gender-neutral natural language queries.  ...  When training a new image search model, the in-processing FairSample method can be used to learn a fairer model from scratch.  ... 
arXiv:2109.05433v1 fatcat:rdthdcdoxvfqhffiuwb65pr3iy

6.1 Knowledge Exchange Consensus: Monitoring of Open Access Publications and Cost Data

Svendsen, Michael; Thomasen, Christian H.
2017 Zenodo  
Purpose and method: In a changing landscape towards increasing OA publishing, it has become necessary for universities and at an aggregated national and international level to monitor OA publications and  ...  clear goal of pushing towards more transparent exchange of metadata of OA and cost data.  ...  More information: Melanie.imming@surfmarket.nl  ... 
doi:10.5281/zenodo.3610213 fatcat:elunpeobxbcwva4ktyb7r62oay

Contrastive Learning for Debiased Candidate Generation in Large-Scale Recommender Systems [article]

Chang Zhou, Jianxin Ma, Jianwei Zhang, Jingren Zhou, Hongxia Yang
2021 arXiv   pre-print
However, live recommender systems face severe exposure bias and have a vocabulary several orders of magnitude larger than that of natural language, implying that MLE will preserve and even exacerbate the  ...  Based on the theoretical discovery, we design CLRec, a contrastive learning method to improve DCG in terms of fairness, effectiveness and efficiency in recommender systems with extremely large candidate  ...  We also find little public information on how a debiased method will eventually affect a live system.  ... 
arXiv:2005.12964v9 fatcat:ju2bph7nqbfhhlvyswuexddova
« Previous Showing results 1 — 15 out of 3,615 results