Filters








417 Hits in 7.6 sec

Multi-Modal Adversarial Autoencoders for Recommendations of Citations and Subject Labels

Lukas Galke, Florian Mai, Iacopo Vagliano, Ansgar Scherp
2018 Proceedings of the 26th Conference on User Modeling, Adaptation and Personalization - UMAP '18  
We analyze the effects of adversarial regularization, sparsity, and different input modalities.  ...  When facing a new recommendation task it is therefore crucial to consider the semantics of item co-occurrence for the choice of an appropriate model.  ...  We have carefully investigated the interaction between the semantics of item co-occurrence and supplying the partial list of items as input for a recommender system.  ... 
doi:10.1145/3209219.3209236 dblp:conf/um/GalkeMVS18 fatcat:yfevueeu2zarjdffri4z2vdyhq

Knowledge Graph Completion: A Review

Zhe Chen, Yuehan Wang, Bin Zhao, Jing Cheng, Xin Zhao, Zongtao Duan
2020 IEEE Access  
Starting from the definition and types of KGC, existing technologies for KGC are analyzed in categories.  ...  Knowledge graph completion (KGC) is a hot topic in knowledge graph construction and related applications, which aims to complete the structure of knowledge graph by predicting the missing entities or relationships  ...  ACKNOWLEDGMENT This work was supported in part by the Funds for Key Research and Development Plan Project of the Shaanxi Province, China, under Grant 2017GY-072, 2019ZDLGY17-08, 2020ZDLGY09-02.  ... 
doi:10.1109/access.2020.3030076 fatcat:jbimdngcmrbx3jhihsgrp62cxq

Learning to Match on Graph for Fashion Compatibility Modeling

Xun Yang, Xiaoyu Du, Meng Wang
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Existing methods have primarily learned visual compatibility from dyadic co-occurrence or co-purchase information of items to model the item-item matching interaction.  ...  Finally, we predict pairwise compatibility based on a compatibility metric learning module. Extensive experiments show that DREP can significantly improve the performance of state-of-the-art methods.  ...  Acknowledgments This research is part of NExT++ research and also supported in part by the National Natural Science Foundation of China (NSFC) under Grant 61725203 and Grant 61732008.  ... 
doi:10.1609/aaai.v34i01.5362 fatcat:rpzqrjyiarbejmidblcznqez4i

Latent Structure Mining with Contrastive Modality Fusion for Multimedia Recommendation [article]

Jinghao Zhang, Yanqiao Zhu, Qiang Liu, Mengqi Zhang, Shu Wu, Liang Wang
2022 arXiv   pre-print
We argue that the latent semantic item-item structures underlying these multimodal contents could be beneficial for learning better item representations and assist the recommender models to comprehensively  ...  Extensive experiments on real-world datasets demonstrate the superiority of our method over state-of-the-art baselines.  ...  Each user u ∈ U is associated with a set of items I u with positive feedbacks which indicate the preference score y ui = 1 for i ∈ I u . x u , x i ∈ R d is the input ID embedding of u and i, respectively  ... 
arXiv:2111.00678v2 fatcat:boqsb2twpjd45gbtol5tpkirqa

Knowledge Extraction And Representation Learning For Music Recommendation And Classification

Sergio Oramas, Xavier Serra
2017 Zenodo  
We focus on the semantic enrichment of descriptions associated to musical items (e.g., artists biographies, album reviews, metadata), and the exploitation of multimodal data (e.g., text, audio, images)  ...  We show how the semantic enrichment of texts and the combination of learned data representations improve the performance on both tasks.  ...  To exploit both the taxonomic and the co-occurrence information, we provide every item with the labels of all their branches.  ... 
doi:10.5281/zenodo.1100973 fatcat:yfpmc6qxbbakjp6qzvywyoaoci

Knowledge Extraction And Representation Learning For Music Recommendation And Classification

Sergio Oramas, Xavier Serra
2017 Zenodo  
We focus on the semantic enrichment of descriptions associated to musical items (e.g., artists biographies, album reviews, metadata), and the exploitation of multimodal data (e.g., text, audio, images)  ...  We show how the semantic enrichment of texts and the combination of learned data representations improve the performance on both tasks.  ...  To exploit both the taxonomic and the co-occurrence information, we provide every item with the labels of all their branches.  ... 
doi:10.5281/zenodo.1048497 fatcat:kdh5jhvocbh3riwln6n2f756su

Recommender systems based on graph embedding techniques: A review

Yue Deng
2022 IEEE Access  
However, in the face of the high complexity and large scale of side information and knowledge, this strategy largely relies for efficient implementation on the scalability of recommendation models.  ...  hidden (indirect) user-item relations, aiming to enrich observed information (or data) for recommendation.  ...  ACKNOWLEDGEMENTS The author acknowledges Linyuan Lü, Shuqi Xu, Xu Na, Hao Wang and Honglei Zhang for their discussions and suggestions.  ... 
doi:10.1109/access.2022.3174197 fatcat:s267xaasovh6ffaomi7l32pqyi

An Enhanced Semantic Layer for Hybrid Recommender Systems

Iván Cantador, Pablo Castells, Alejandro Bellogín
2011 International Journal on Semantic Web and Information Systems (IJSWIS)  
Challenging issues in their research agenda include the sparsity of user preference data, and the lack of flexibility to incorporate contextual factors in the recommendation methods.  ...  The enhanced semantics support the development of contextualisation capabilities, and enable performance improvements in recommendation methods.  ...  ones (synonyms, co-occurrences, etc.).  ... 
doi:10.4018/jswis.2011010103 fatcat:q7nptioe6zawzi7de7dj3dezcm

Methodologies for Cross-Domain Data Fusion: An Overview

Yu Zheng
2015 IEEE Transactions on Big Data  
Traditional data mining usually deals with data from a single domain. In the big data era, we face a diversity of datasets from different sources in different domains.  ...  These datasets consist of multiple modalities, each of which has a different representation, distribution, scale, and density.  ...  If this location-activity matrix is completely filled, we can recommend a set of locations for a particular activity by retrieving the top k locations with a relatively high frequency from the column that  ... 
doi:10.1109/tbdata.2015.2465959 fatcat:flm37ozmhzcrfbrzeuagxm4l6a

Explainable Outfit Recommendation with Joint Outfit Matching and Comment Generation

Yujie Lin, Pengjie Ren, Zhumin Chen, Zhaochun Ren, Jun Ma, Maarten De Rijke
2019 IEEE Transactions on Knowledge and Data Engineering  
Extensive experiments conducted on an existing dataset and a collected real-world dataset show NOR achieves significant improvements over state-of-the-art baselines for outfit recommendation.  ...  Existing work neglects user comments of fashion items, which have been proved to be effective in generating explanations along with better recommendation results.  ...  , Ahold Delhaize, the Association of Universities in the Netherlands, and the Innovation Center for Artificial Intelligence (ICAI).  ... 
doi:10.1109/tkde.2019.2906190 fatcat:rxnzj2u7ebbdzo6pjfnfilhb3i

Explainable Outfit Recommendation with Joint Outfit Matching and Comment Generation [article]

Yujie Lin, Pengjie Ren, Zhumin Chen, Zhaochun Ren, Jun Ma, Maarten de Rijke
2019 arXiv   pre-print
Extensive experiments conducted on an existing dataset and a collected real-world dataset show NOR achieves significant improvements over state-of-the-art baselines for outfit recommendation.  ...  Existing work neglects user comments of fashion items, which have been proved to be effective in generating explanations along with better recommendation results.  ...  , Ahold Delhaize, the Association of Universities in the Netherlands, and the Innovation Center for Artificial Intelligence (ICAI).  ... 
arXiv:1806.08977v3 fatcat:u6uqgnhy6nadpgn3wknkpsirym

User Response Prediction in Online Advertising [article]

Zhabiz Gharibshah, Xingquan Zhu
2021 arXiv   pre-print
What are the parties involved in the online digital advertising eco-systems? What type of data are available for user response prediction?  ...  In this survey, we provide a comprehensive review of user response prediction in online advertising and related recommender applications.  ...  A study [71] showed that sequence of user behavior can be organized as a co-occurrence commodity graph with node representing clicked commodities and weighted edges describing number of co-occurrence  ... 
arXiv:2101.02342v2 fatcat:clgefamcd5fmbeg5ephizy3zqu

A Hybrid Probabilistic Model for Unified Collaborative and Content-Based Image Tagging

Ning Zhou, W K Cheung, Guoping Qiu, Xiangyang Xue
2011 IEEE Transactions on Pattern Analysis and Machine Intelligence  
We developed a collaborative filtering method based on non-negative matrix factorization (NMF) for tackling this data sparsity issue.  ...  For images with user provided tags, HPM jointly exploits both the image features and the tags in a unified probability framework to recommend additional tags to label the images.  ...  used for estimating tag co-occurrence probabilities in the HPM for the Corel5k data set.  ... 
doi:10.1109/tpami.2010.204 pmid:21079279 fatcat:jerygw4q7bhlfbldue7vsoyk4i

Survey for Trust-aware Recommender Systems: A Deep Learning Perspective [article]

Manqing Dong, Feng Yuan, Lina Yao, Xianzhi Wang, Xiwei Xu, Liming Zhu
2020 arXiv   pre-print
A significant remaining challenge for existing recommender systems is that users may not trust the recommender systems for either lack of explanation or inaccurate recommendation results.  ...  We focus on the work based on deep learning techniques, an emerging area in the recommendation research.  ...  Some may construct the graph for items to model the user behaviors: each item is represented by a node in the graph, and the co-occurrence of items is denoted by edge.  ... 
arXiv:2004.03774v2 fatcat:q7mehir7hbbzpemw3q5fkby5ty

Inferring Semantic Relations by User Feedback [chapter]

Francesco Osborne, Enrico Motta
2014 Lecture Notes in Computer Science  
To address these issues, we propose Klink UM, an approach for extracting emergent semantics from user feedbacks, with the aim of tailoring the ontology to the users and improving the recommendations accuracy  ...  In the last ten years, ontology-based recommender systems have been shown to be effective tools for predicting user preferences and suggesting items.  ...  We ran Klink UM and the baseline method 10 times for each different set of randomized input data and compared the generated ontologies with the two original gold standard ontologies, using the average  ... 
doi:10.1007/978-3-319-13704-9_27 fatcat:o5fafyoq3ngclcktvu7zrbtlp4
« Previous Showing results 1 — 15 out of 417 results