Filters








33 Hits in 3.5 sec

PBODL : Parallel Bayesian Online Deep Learning for Click-Through Rate Prediction in Tencent Advertising System [article]

Xun Liu, Wei Xue, Lei Xiao, Bo Zhang
2017 arXiv   pre-print
MatchBox.  ...  Then we extend the model family to a variety of bayesian online models with increasing feature embedding capabilities, such as Sparse-MLP, FM-MLP and FFM-MLP.  ...  Considering the scale of the learning problem, online learning (bayesian online learning in particular) paradigm is chosen as a corner stone of our solution. • Nonlinear Model : Linear models depends heavily  ... 
arXiv:1707.00802v2 fatcat:lab2rbetdbegxjedkzi32sxzfu

The Xbox recommender system

Noam Koenigstein, Nir Nice, Ulrich Paquet, Nir Schleyen
2012 Proceedings of the sixth ACM conference on Recommender systems - RecSys '12  
The system largely relies on implicit feedback, and runs on a large scale, serving tens of millions of daily users. We describe the system design, and review the core recommendation algorithm.  ...  A recent addition to Microsoft's Xbox Live Marketplace is a recommender system which allows users to explore both movies and games in a personalized context.  ...  In this paper we wish to bring a complete overview of a real world large scale recommender system.  ... 
doi:10.1145/2365952.2366015 dblp:conf/recsys/KoenigsteinNPS12 fatcat:eo3nkduc35f2jddjrmgptf3c4u

A Distributed Collaborative Filtering Algorithm Using Multiple Data Sources [article]

Mohamed Reda Bouadjenek, Esther Pacitti, Maximilien Servajean, Florent Masseglia, Amr El Abbadi
2018 arXiv   pre-print
Collaborative Filtering (CF) is one of the most commonly used recommendation methods.  ...  These valuable data sources may supply useful information to enhance a recommendation system in modeling users' preferences and item characteristics more accurately and thus, hopefully, to make recommenders  ...  to show that it can scale to large datasets.  ... 
arXiv:1807.05853v1 fatcat:uh3slasdvrghhb7jp3dr5u5jqi

Mining large streams of user data for personalized recommendations

Xavier Amatriain
2013 SIGKDD Explorations  
But since then, Recommender Systems have evolved.  ...  I then describe the use of recommendation and personalization techniques at Netflix.  ...  INTRODUCTION Recommender Systems (RS) are a prime example of the mainstream applicability of large scale data mining.  ... 
doi:10.1145/2481244.2481250 fatcat:5h35wzeaufdltaq4ckyufaayjy

CoBayes

Gjergji Kasneci, Jurgen Van Gael, David Stern, Thore Graepel
2011 Proceedings of the fourth ACM international conference on Web search and data mining - WSDM '11  
Our work aims at building probabilistic tools for constructing and maintaining large-scale knowledge bases containing entity-relationship-entity triples (statements) extracted from the Web.  ...  Bayesian inference in this complex graphical model is performed using mixed variational and expectation propagation message passing.  ...  The expertise model takes the form of a recommendation system like Matchbox [22] .  ... 
doi:10.1145/1935826.1935896 dblp:conf/wsdm/KasneciGSG11 fatcat:fxgky6aabjg4ffzrrrnh3hss5q

A hierarchical model for ordinal matrix factorization

Ulrich Paquet, Blaise Thomson, Ole Winther
2011 Statistics and computing  
Importantly, these algorithms may be implemented in the factorization of very large matrices with missing entries.  ...  Results also show how Gibbs sampling outperforms variational Bayes on this task, despite the large number of ratings and model parameters.  ...  A further benefit of a Bayesian approach is that recommendations include a predictive variance. Practical systems can exploit this to only recommend items that it is confident a user will like.  ... 
doi:10.1007/s11222-011-9264-x fatcat:kqxklkxsrrck3jknga3znlfrx4

Collaborative personalized tweet recommendation

Kailong Chen, Tianqi Chen, Guoqing Zheng, Ou Jin, Enpeng Yao, Yong Yu
2012 Proceedings of the 35th international ACM SIGIR conference on Research and development in information retrieval - SIGIR '12  
Twitter has rapidly grown to a popular social network in recent years and provides a large number of real-time messages for users.  ...  In this paper, we focus on recommending useful tweets that users are really interested in personally to reduce the users' effort to find useful information.  ...  Zaman [35] uses Matchbox, a large scale online Bayesian recommendation method to predict retweets.  ... 
doi:10.1145/2348283.2348372 dblp:conf/sigir/ChenCZJYY12 fatcat:iimquokm3rdnvetvo7fnkwa2dq

Conditional Expectation Propagation [article]

Zheng Wang, Shandian Zhe
2019 arXiv   pre-print
Matchbox: large scale online Bayesian recommendations. In Proceedings of the 18th international conference on World wide web, pages 111-120. ACM. Zhe, S., Lee, K.  ...  This is important for large-scale applications.  ... 
arXiv:1910.12360v2 fatcat:xqauglrp6vacxghutmpoqz2pbe

Attribute-aware Collaborative Filtering: Survey and Classification [article]

Wen-Hao Chen, Chin-Chi Hsu, Yi-An Lai, Vincent Liu, Mi-Yen Yeh, Shou-De Lin
2018 arXiv   pre-print
ributes (if the original a ribute contains d categories, the dimension of transformed a ributes would be d), we nd that most of the experimented baseline models cannot nish training in hours for some large-scale  ...  Bayesian Personalized Ranking (BPR) [82] rst investigates the usage and the optimization of (22) for recommender systems.  ... 
arXiv:1810.08765v1 fatcat:gkjhxziqengxbog5greaziponq

Attribute-Aware Recommender System Based on Collaborative Filtering: Survey and Classification

Wen-Hao Chen, Chin-Chi Hsu, Yi-An Lai, Vincent Liu, Mi-Yen Yeh, Shou-De Lin
2020 Frontiers in Big Data  
webscope.sandbox.yahoo.com/ 3 https://www.yelp.com/dataset significantly increases the dimensions of attributes, we find that many of the experimented models cannot finish training in a reasonable amount of time for large-scale  ...  Bayesian Personalized Ranking (BPR) (Rendle et al., 2009) first investigated the usage and the optimization of (20) for recommender systems.  ... 
doi:10.3389/fdata.2019.00049 pmid:33693372 pmcid:PMC7931907 fatcat:yt4l4mp2hfd47do5opvyf73kyy

Digital Wildfires

Helena Webb, Pete Burnap, Rob Procter, Omer Rana, Bernd Carsten Stahl, Matthew Williams, William Housley, Adam Edwards, Marina Jirotka
2016 ACM Transactions on Information Systems  
Social media platforms provide an increasingly popular means for individuals to share content online.  ...  They identified a Bayesian logistic regression as a favoured predictive model.  ...  [Zaman et al. 2010 ] used the MatchBox algorithm to predict retweet probability for individual tweets, finding that attributes of the tweeter and the retweeter (similar to author profile of Suh et al.  ... 
doi:10.1145/2893478 fatcat:7tmgaiqdufgudf3t2jwq6savay

Automated feature generation from structured knowledge

Weiwei Cheng, Gjergji Kasneci, Thore Graepel, David Stern, Ralf Herbrich
2011 Proceedings of the 20th ACM international conference on Information and knowledge management - CIKM '11  
The model we use is a generalized linear Bayesian probit model, similar to the ones used in Matchbox [23] or AdPredictor [25] .  ...  However, for more and more large-scale learning problems that are typical for web-scale applications of machine learning, the situation is different.  ... 
doi:10.1145/2063576.2063779 dblp:conf/cikm/ChengKGSH11 fatcat:skherxpctfd4bgfipgqbpin4b4

Bayesian recommender systems : models and algorithms [article]

Shengbo Guo, University, The Australian National, University, The Australian National
2018
First, we present compact Bayesian graphical models for dimensions (1)-(3). Second, for each dimension, we make use of advanced Bayesian inference techniques to learn and make optimal recommendations.  ...  This thesis is about how Bayesian methods can be applied to explicitly model and efficiently reason about uncertainty to make optimal recommendations.  ...  Stern et al. (2009) make use of user and item meta data, and propose a large scale online Bayesian Non-Bayesian vs.  ... 
doi:10.25911/5d500072c24b0 fatcat:4rjhdux7rrb65hqeof4lhrnhxa

A Unified Approach to Collaborative Filtering via Linear Models and Beyond [article]

Suvash Sedhain, University, The Australian National, University, The Australian National
2017
We formulate a large-scale linear model that leverages users social information.  ...  Recommending a personalised list of items to users is a core task for many online services such as Amazon, Netflix, and Youtube.  ...  CBF makes a recommendation leveraging user and item  ... 
doi:10.25911/5d6fa25393880 fatcat:cvqfhflooncrhijmy3rfy5226y

The nuclear and mitochondrial genomes of Frieseomelitta varia – a highly eusocial stingless bee (Meliponini) with a permanently sterile worker caste

Flávia C. de Paula Freitas, Anete P. Lourenço, Francis M. F. Nunes, Alexandre R. Paschoal, Fabiano C. P. Abreu, Fábio O. Barbin, Luana Bataglia, Carlos A. M. Cardoso-Júnior, Mário S. Cervoni, Saura R. Silva, Fernanda Dalarmi, Marco A. Del Lama (+19 others)
2020 BMC Genomics  
We also predicted 169,371 repetitive genomic components, 2083 putative transposable elements, and 1946 genes for non-coding RNAs, largely long non-coding RNAs.  ...  Using specific prediction methods, we identified a large number of repetitive genome components and long non-coding RNAs, which could provide the molecular basis for gene regulatory plasticity, including  ...  For instance, while colonies of the tiny, fruit fly-size Leurotrigona species can fit into a matchbox, colonies of the opennesting Trigona species can be of a size comparable to that of very large honey  ... 
doi:10.1186/s12864-020-06784-8 pmid:32493270 fatcat:xxnsfspn7vcdrgg5cpfrh5z7h4
« Previous Showing results 1 — 15 out of 33 results