Filters








116,372 Hits in 3.2 sec

Explanation-Guided Fairness Testing through Genetic Algorithm [article]

Ming Fan, Wenying Wei, Wuxia Jin, Zijiang Yang, Ting Liu
2022 arXiv   pre-print
The fairness characteristic is a critical attribute of trusted AI systems. A plethora of research has proposed diverse methods for individual fairness testing.  ...  We first rank the explanation result 𝑒 based on the importance score assigned by the interpretable method.  ...  Our experiments are conducted on a server with Ubuntu 18.04 operating system, Intel Xeon 2.50GHz CPU, NVIDIA RTX GPU, and 128GB system memory.  ... 
arXiv:2205.08335v1 fatcat:kwcxbsoif5ct3cq4m4i77rwee4

From chatter to headlines

Gianmarco De Francisci Morales, Aristides Gionis, Claudio Lucchese
2012 Proceedings of the fifth ACM international conference on Web search and data mining - WSDM '12  
We propose a new methodology for recommending interesting news to users by exploiting the information in their twitter persona.  ...  stream, and topic popularity in the news and in the whole twitter-land.  ...  Our main recommendation system is called T.Rex, for twitter-based news recommendation system. • Our system provides personalized recommendations by leveraging information from the tweet stream of users  ... 
doi:10.1145/2124295.2124315 dblp:conf/wsdm/MoralesGL12 fatcat:6bkgdiirkbceziko243frivooa

Real-time top-n recommendation in social streams

Ernesto Diaz-Aviles, Lucas Drumond, Lars Schmidt-Thieme, Wolfgang Nejdl
2012 Proceedings of the sixth ACM conference on Recommender systems - RecSys '12  
In this paper, we focus on analyzing social streams in real-time for personalized topic recommendation and discovery.  ...  We consider collaborative filtering as an online ranking problem and present Stream Ranking Matrix Factorization -RMFX -, which uses a pairwise approach to matrix factorization in order to optimize the  ...  The experiments were conducted using GNU/Linux 64-bit as operating system. None of the methods was parallelized and therefore used one single CPU for computations.  ... 
doi:10.1145/2365952.2365968 dblp:conf/recsys/Diaz-AvilesDSN12 fatcat:pzemwfmflndn5iqe245p6mwpru

Towards real-time collaborative filtering for big fast data

Ernesto Diaz-Aviles, Wolfgang Nejdl, Lucas Drumond, Lars Schmidt-Thieme
2013 Proceedings of the 22nd International Conference on World Wide Web - WWW '13 Companion  
However, given the deluge of data items, it is a challenge for individuals to find relevant and appropriately ranked information at the right time.  ...  , towards an individual user perspective, and ask: "What is interesting to me right now within the social media stream?".  ...  CONCLUSION Our research on online collaborative filtering for social media streams provides an example of integrating large-scale recommender systems with the real-time nature of Twitter.  ... 
doi:10.1145/2487788.2488044 dblp:conf/www/Diaz-AvilesNDS13 fatcat:tqvscm56afhwteilaeuso5djey

Parallel methods for evidence and trust based selection and recommendation of software apps from online marketplaces

Lahiru S. Gallege, Rajeev R. Raje
2017 Proceedings of the 12th Annual Conference on Cyber and Information Security Research - CISRC '17  
In addition to feature-based information, about these apps, these marketplaces contain large volumes of user reviews.  ...  CCS CONCEPTS • Information systems ~ Trust • Information systems R ecommender systems • Social and professional topics ~ Software selection and adaptation  ...  [22] experiments on a recommender system to parallelize data using Hadoop echo-system. It collects data from users, commodities, and transactions.  ... 
doi:10.1145/3064814.3064819 dblp:conf/csiirw/GallegeR17 fatcat:3a32egf2ijckvkw7s4dmxhi7ru

What is happening right now ... that interests me?

Ernesto Diaz-Aviles, Lucas Drumond, Zeno Gantner, Lars Schmidt-Thieme, Wolfgang Nejdl
2012 Proceedings of the 21st ACM international conference on Information and knowledge management - CIKM '12  
In this paper, we consider collaborative filtering as an online ranking problem and present RMFO, a method that creates, in real-time, user-specific rankings for a set of tweets based on individual preferences  ...  Experiments on the 476 million Twitter tweets dataset show that our online approach largely outperforms recommendations based on Twitter's global trend and Weighted Regularized Matrix Factorization (WRMF  ...  We introduce a novel framework for online collaborative filtering based on a pairwise ranking approach for matrix factorization, in the presence of streaming data. 2.  ... 
doi:10.1145/2396761.2398479 dblp:conf/cikm/Diaz-AvilesDGSN12 fatcat:bwbbyygj6jcabpypet7cp6hw7m

Performance Improvement of Stream-Centered Probabilistic Matrix Factorization Method Using Weighted Reservoir Sampling and Parallel Computing

Berman Danyel Sinaga, School of Electrical Engineering and Informatics, ITB, Indonesia Ganesha 10, Bandung 40132, Indonesia, Saiful Akbar, School of Electrical Engineering and Informatics, ITB, Indonesia Ganesha 10, Bandung 40132, Indonesia
2021 International Journal on Electrical Engineering and Informatics  
Application of the recommender system on the online platform turns out to produce a large volume of data and at an unexpected rate, making it more realistic to study the recommender system under streaming  ...  Many studies have tried to develop a streaming recommender system, one of which is Stream-centered Probabilistic Matrix Factorization (SPMF).  ...  Acknowledgement filtering based on a data stream management system," CEUR Workshop Proc., vol.  ... 
doi:10.15676/ijeei.2021.13.3.8 fatcat:kb4eyrcejrhpxgpanrarzfhuby

Guest editorial: web information technologies

Xuemin Lin, Jeffrey Xu Yu
2015 World wide web (Bussum)  
The second paper, by Li et al., "Social event identification and ranking on flickr", focuses on social event modeling and ranking.  ...  Furthermore, event impact is defined and estimated via random walk based on the triggering relationships.  ...  Yao et al. address the limitations of recommendation systems in existing techniques and cope with scenarios with only implicit feedback in the seventh paper, "A Graph-based Model for Context-aware Recommendation  ... 
doi:10.1007/s11280-015-0356-y fatcat:uulmqqc4sneh3mlaqw3zlrpmly

Considering Durations and Replays to Improve Music Recommender Systems [article]

Pierre Hanna
2017 arXiv   pre-print
A quantitative study as usually found in the literature confirms that neighborhood-based systems considering binary data give the best results in terms of MAP@k.  ...  A large database was thus created using logs collected on a streaming platform, notably collecting the listening times.  ...  Acknowledgements The author wishes to thank Simbals team, Deezer R&D and Recommendation teams for making this work possible, in particular Manuel Moussalam, Thomas Bouabca and Aurélien Hérault.  ... 
arXiv:1711.05237v1 fatcat:5t3xoh5cp5gpzaf4wqhb2mciay

Considering Durations and Replays to Improve Music Recommender Systems

Pierre Hanna
2018 EAI Endorsed Transactions on Self-Adaptive Systems  
A quantitative study as usually found in the literature confirms that neighborhood-based systems considering binary data give good results in terms of MAP@k.  ...  A large database was thus created using logs collected on a streaming platform, notably collecting the listening times.  ...  Acknowledgements The author wishes to thank Simbals team, Deezer R&D and Recommendation teams for making this work possible, in particular Manuel Moussalam, Thomas Bouabca and Aurélien Hérault.  ... 
doi:10.4108/eai.5-2-2018.156379 fatcat:hbb4ojxk4fhm7ebdsnsiuay724

Tutorial on Open Source Online Learning Recommenders

Róbert Pálovics, Domokos Kelen, András A. Benczúr
2017 Proceedings of the Eleventh ACM Conference on Recommender Systems - RecSys '17  
In our tutorial, we present open source systems capable of updating their models on the fly after each event: Apache Spark, Apache Flink and Alpenglow, a C++ based Python recommender framework.  ...  Recommender systems have to serve in online environments that can be non-stationary.  ...  An important part of the tutorial considers the framework for evaluating recommender systems over streaming data. We rely on ideas of [1, 7] for the online DCG measure.  ... 
doi:10.1145/3109859.3109937 dblp:conf/recsys/PalovicsKB17 fatcat:7nigretoqncl3bxv4tc2gs2hgm

A recommender system architecture for predictive telecom network management

Faisal Zaman, Gabriel Hogan, Sven Der Meer, John Keeney, Sebastian Robitzsch, Gabriel-miro Muntean
2015 IEEE Communications Magazine  
This work presents the design and specification of E-Stream, a predictive recommendation based solution to automated network management.  ...  After observing event sequences in incoming event streams, specific appropriate actions are selected, ranked and recommended to pre-empt the predicted incidents.  ...  E-STREAM COMPONENTS E-stream is a composite system to recommend corrective solutions to network scenarios based on predictive patterns leading up to network incidences.  ... 
doi:10.1109/mcom.2015.7010547 fatcat:36fg4bxfofbfpjz5hbdazhe42m

Short and tweet

Jilin Chen, Rowan Nairn, Les Nelson, Michael Bernstein, Ed Chi
2010 Proceedings of the 28th international conference on Human factors in computing systems - CHI '10  
We conclude this work by discussing the implications of our recommender design and how our design can generalize to other information streams.  ...  More and more web users keep up with newest information through information streams such as the popular microblogging website Twitter.  ...  Andersen et al. discussed several key insights in their theory of trust-based recommender systems [2] , one of which is trust propagation.  ... 
doi:10.1145/1753326.1753503 dblp:conf/chi/ChenNNBC10 fatcat:ab3qj5jmo5azfgqhxfeebysjv4

SciRecSys: A Recommendation System for Scientific Publication by Discovering Keyword Relationships [chapter]

Vu Le Anh, Vo Hoang Hai, Hung Nghiep Tran, Jason J. Jung
2014 Lecture Notes in Computer Science  
Particularly, a recommendation system (called SciRecSys) has been presented to support users to efficiently find out relevant articles.  ...  In this work, we propose a new approach for discovering various relationships among keywords over the scientific publications based on a Markov Chain model.  ...  Processing Data Stream 2 Data Stream Stream Processing 3 Database System Database System 4 Sensor Network Object Oriented 5 Query Processing Data Model (b) D2 in database domain, |K| = 2755  ... 
doi:10.1007/978-3-319-11289-3_8 fatcat:wjwe7wl5rfepports423lprb2i

Recommender Systems Over Data Streams [chapter]

András A. Benczúr, Levente Kocsis, Róbert Pálovics
2018 Encyclopedia of Big Data Technologies  
A practical recommender system displays a ranked list of a few items for which the user can give feedback.  ...  In this section, we show the main differences in evaluating such systems compared to both classifiers and batch systems, as well as describe the main data stream recommender algorithms.  ... 
doi:10.1007/978-3-319-63962-8_328-1 fatcat:bleg7iemrrgxpbl3qlrrq2msoy
« Previous Showing results 1 — 15 out of 116,372 results