263 Hits in 4.1 sec

Connecting comments and tags

Dawei Yin, Shengbo Guo, Boris Chidlovskii, Brian D. Davison, Cedric Archambeau, Guillaume Bouchard
2013 Proceedings of the sixth ACM international conference on Web search and data mining - WSDM '13  
We tackle these problems by using a generalized latent factor model and fully Bayesian treatment. To evaluate performance, we test on two real-world data sets from Flickr and Bibsonomy.  ...  In this paper, we investigate improved modeling for predictive social tagging systems.  ...  One related method is collective matrix factorization from Singh and Gordon [31] which provides a general framework to model multi-relational data, extending many previous approaches for matrix factorization  ... 
doi:10.1145/2433396.2433466 dblp:conf/wsdm/YinGC0AB13 fatcat:t7u7hvft5fapdkp5ugupps7wki

BASIL: Effective Near-Duplicate Image Detection Using Gene Sequence Alignment [chapter]

Hung-sik Kim, Hau-Wen Chang, Jeongkyu Lee, Dongwon Lee
2010 Lecture Notes in Computer Science  
Based on the same characteristics principle, we propose a matrix co-factorization framework with a shared latent space to optimize the recommendation collectively.  ...  First, we study the home location estimation for Twitter users from their shared textual content. We employ Gaussian Mixture Model to compensate the drawback in the Maximum Likelihood Estimation.  ...  R UG , R UI and R GI is the rating matrix for user-group, user-image and group-images, respectively. • Weighting Matrix W . The matrix used to develop Weighted Regularized Matrix Factorization.  ... 
doi:10.1007/978-3-642-12275-0_22 fatcat:ou4wo4a6efdabkipzbkaxd5cyi

Automatic image semantic interpretation using social action and tagging data

Neela Sawant, Jia Li, James Z. Wang
2010 Multimedia tools and applications  
Social inputs with their added context represent a strong substitute for expert annotations.  ...  The survey concludes with a summary of principle research directions for the present and the future.  ...  The prestige of special interest groups in which the photo appears is also a contributing factor [49] .  ... 
doi:10.1007/s11042-010-0650-8 fatcat:kqu6kyess5f3re554jsueuuzem

A sparse gaussian processes classification framework for fast tag suggestions

Yang Song, Lu Zhang, C. Lee Giles
2008 Proceeding of the 17th ACM conference on Information and knowledge mining - CIKM '08  
We suggest a novel prototype selection algorithm to select the best subset of points for model learning.  ...  In addition, results on tagging digital data also demonstrate that our model is capable of recommending relevant tags to images and videos by using their surrounding textual information.  ...  Keywords are then extracted from similar documents for recommendation.  ... 
doi:10.1145/1458082.1458098 dblp:conf/cikm/SongZG08 fatcat:f3fjlegs4nelnml4hopdtfqmdu

Affective Image Content Analysis: Two Decades Review and New Perspectives [article]

Sicheng Zhao, Xingxu Yao, Jufeng Yang, Guoli Jia, Guiguang Ding, Tat-Seng Chua, Björn W. Schuller, Kurt Keutzer
2021 arXiv   pre-print
Finally, we discuss some challenges and promising research directions in the future, such as image content and context understanding, group emotion clustering, and viewer-image interaction.  ...  We begin with an introduction to the key emotion representation models that have been widely employed in AICA and description of available datasets for performing evaluation with quantitative comparison  ...  Affective image analysis for groups of people, i.e. recognizing the induced emotions of the groups, has not been explored yet. Group emotion recognition plays an important role in recommendation.  ... 
arXiv:2106.16125v1 fatcat:5y5y5nhoebccxjjybarnveecgq

Multilabel user classification using the community structure of online networks

Georgios Rizos, Symeon Papadopoulos, Yiannis Kompatsiaris, Yudong Zhang
2017 PLoS ONE  
Then, we perform supervised community feature weighting in order to boost the importance of highly predictive communities.  ...  To this end, we employ an improvement of personalized PageRank algorithms for searching locally in each user's graph structure.  ...  It has successfully found application in personalized content and user recommendation [1] , targeted advertising and marketing, expert search and monitoring [2] and social search and promotion [3,  ... 
doi:10.1371/journal.pone.0173347 pmid:28278242 pmcid:PMC5344401 fatcat:evelzyqlofgapgjf7nhnt44aea

Socializing the Semantic Gap

Xirong Li, Tiberio Uricchio, Lamberto Ballan, Marco Bertini, Cees G. M. Snoek, Alberto Del Bimbo
2016 ACM Computing Surveys  
by various research groups.  ...  Putting all this together, the survey aims to provide an overview of the past and foster progress for the near future.  ...  Jitao Sang for providing the TensorAnalysis results, and Dr. Meng Wang and Dr. Yue Gao for making the Flickr51 dataset available for this survey.  ... 
doi:10.1145/2906152 fatcat:ir6obqydirgihlu54pnwy4clny

Analyzing the Dynamics of Communication in Online Social Networks [chapter]

Munmun De Choudhury, Hari Sundaram, Ajita John, Doree Duncan Seligmann
2010 Handbook of Social Network Technologies and Applications  
My central hypothesis is that communication dynamics between individuals manifest themselves via three key aspects: the information that is the content of communication, the social engagement i.e. the  ...  this disseration is essentially computational as well as empirical: I present characterization techniques, develop computational models and finally discuss large-scale quantitative observational studies for  ...  Flickr Dataset We have tested our group recommendation framework on a dataset crawled from the popular media-sharing site, Flickr.  ... 
doi:10.1007/978-1-4419-7142-5_4 fatcat:d2mn6rm64zg2bdgo4kdoysyi64

Social tag relevance learning via ranking-oriented neighbor voting

Chaoran Cui, Jialie Shen, Jun Ma, Tao Lian
2016 Multimedia tools and applications  
However, user-generated tags in real world are often imprecise and incomplete to describe the image contents, which severely degrades the performance of current search systems.  ...  To improve the descriptive powers of social tags, a fundamental issue is tag relevance learning, which concerns how to interpret the relevance of a tag with respect to the contents of an image effectively  ...  Note that ξ is the point-wise maximum of a set {ξ 1 , ξ 2 , . . .}, where ξ i is the margin for the ith constraint in W.  ... 
doi:10.1007/s11042-016-3512-1 fatcat:oxwe4wikgff2tcpccvg3vchi5u

Recurrent Coupled Topic Modeling over Sequential Documents [article]

Jinjin Guo, Longbing Cao, Zhiguo Gong
2021 arXiv   pre-print
In addition, the latent Indian Buffet Process (IBP) compound distribution is exploited to automatically infer the overall topic number and customize the sparse topic proportions for each sequential document  ...  Dynamic Poisson factor analysis. Targeting at the count data, it is a matrix factorization method for the discrete sequential count data under the Poisson factor analysis (PFA) [1] .  ...  Sampling : the sparse document-topic affinity matrix could be sampled by marginalizing out and .  ... 
arXiv:2106.13732v1 fatcat:hbfxszcnevgfplgnxiiqfnturm

Shilling attack detection for recommender systems based on credibility of group users and rating time series

Wei Zhou, Junhao Wen, Qiang Qu, Jun Zeng, Tian Cheng, Hua Wang
2018 PLoS ONE  
Recommender systems are vulnerable to shilling attacks. Forged user-generated content data, such as user ratings and reviews, are used by attackers to manipulate recommendation rankings.  ...  Standard testing datasets are used to verify the effect of the proposed method. (2018) Shilling attack detection for recommender systems based on credibility of group users and rating time series.  ...  Recommender systems can be divided into two categories, collaborative filtering and content-based filtering.  ... 
doi:10.1371/journal.pone.0196533 pmid:29742134 pmcid:PMC5942815 fatcat:ckuiplr7zjcsxjyccmnfscggci

Adversarial Learning for Personalized Tag Recommendation [article]

Erik Quintanilla, Yogesh Rawat, Andrey Sakryukin, Mubarak Shah, Mohan Kankanhalli
2020 arXiv   pre-print
A joint training of user-preference and visual encoding allows the network to efficiently integrate the visual preference with tagging behavior for a better user recommendation.  ...  In this paper, we address the problem of personalized tag recommendation and propose an end-to-end deep network which can be trained on large-scale datasets.  ...  Most of the existing methods for personalized tag recommendation make use of tag cooccurrence [30] , matrix factorization [31] , [32] , [33] , and graph-based models [34] .  ... 
arXiv:2004.00698v1 fatcat:on4mybzeyfdc3gvnndedz4hpma

Learning Compact Visual Descriptors for Low Bit Rate Mobile Landmark Search

Ling-Yu Duan, Jie Chen, Rongrong Ji, Tiejun Huang, Wen Gao
2013 The AI Magazine  
Our descriptor outperforms alternative compact descriptors (Chen et al. 2009; Chen et al., 2010; Chandrasekhar et al. 2009a; Chandrasekhar et al. 2009b) with significant margins.  ...  The compactness originates from the bag-of-words image representation, with an offline learning from geotagged photos from online photo sharing websites including Flickr and Panoramio.  ...  recommendation (Hays and Efros 2003; Li et al. 2008; Zheng et al. 2009 ), and content-based advertising (Liu et al. 2009 ).  ... 
doi:10.1609/aimag.v34i2.2469 fatcat:a26mz5iqs5dwtbuwdovlwaeoca

Landmark Classification With Hierarchical Multi-Modal Exemplar Feature

Lei Zhu, Jialie Shen, Hai Jin, Liang Xie, Ran Zheng
2015 IEEE transactions on multimedia  
Landmark image classification attracts increasing research attention due to its great importance in real applications, ranging from travel guide recommendation to 3-D modelling and visualization of geolocation  ...  Then, at the stage of exemplar selection, hierarchical discriminative exemplars in multiple modalities are discovered automatically via iterative boosting and latent region label mining.  ...  ACKNOWLEDGMENT The authors would like to thank the anonymous reviewers for their constructive and helpful suggestions.  ... 
doi:10.1109/tmm.2015.2431496 fatcat:hhyhhp2ctzbzznm5sfidiiv7ey

Content is still king

Ba Quan Truong, Aixin Sun, Sourav S. Bhowmick
2012 Proceedings of the 2nd ACM International Conference on Multimedia Retrieval - ICMR '12  
relevance learning for TagIR.  ...  images) and contextual (e.g., tag association by co-occurrence) features of social images to further boost the accuracy of TagIR.  ...  of a group of correlated tags further boost the accuracy of tag relevance computation as these tags jointly describe the visual concept of the image.  ... 
doi:10.1145/2324796.2324808 dblp:conf/mir/TruongSB12 fatcat:ydlxsnxrdvegzcqafgvcghgery
« Previous Showing results 1 — 15 out of 263 results