Filters








3,323 Hits in 5.4 sec

MetaCVR: Conversion Rate Prediction via Meta Learning in Small-Scale Recommendation Scenarios [article]

Xiaofeng Pan, Ming Li, Jing Zhang, Keren Yu, Luping Wang, Hong Wen, Chengjun Mao, Bo Cao
2022 arXiv   pre-print
In this work, we propose a novel CVR method named MetaCVR from a perspective of meta learning to address the DDF issue.  ...  To the best of our knowledge, this is the first study of CVR prediction targeting the DDF issue in small-scale recommendation scenarios.  ...  Item ID, category ID and brand ID have an embedding size of 32 while 8 for the other categorical features. We use 8-head attention structures with a hidden size of 128.  ... 
arXiv:2112.13753v4 fatcat:u3st3d3rhrc2xm7arvz5d3qwfu

Zero-Shot Recommender Systems [article]

Hao Ding, Yifei Ma, Anoop Deoras, Yuyang Wang, Hao Wang
2021 arXiv   pre-print
Different from categorical item indices, i.e., item ID, in previous methods, ZESRec uses items' natural-language descriptions (or description embeddings) as their continuous indices, and therefore naturally  ...  We study three pairs of real-world RS datasets and demonstrate that ZESRec can successfully enable recommendations in such a zero-shot setting, opening up new opportunities for resolving the chicken-and-egg  ...  We also consider their extensions, HRNN-Meta, GRU4Rec-Meta, and TCN-Meta, which use items' NL description embeddings to replace item ID hidden embeddings.  ... 
arXiv:2105.08318v2 fatcat:tbpnyiaefrhnlnlwsjqj3qlmbu

CAN: Feature Co-Action for Click-Through Rate Prediction [article]

Weijie Bian, Kailun Wu, Lejian Ren, Qi Pi, Yujing Zhang, Can Xiao, Xiang-Rong Sheng, Yong-Nan Zhu, Zhangming Chan, Na Mou, Xinchen Luo, Shiming Xiang (+3 others)
2021 arXiv   pre-print
Feature interaction has been recognized as an important problem in machine learning, which is also very essential for click-through rate (CTR) prediction tasks.  ...  More specifically, giving feature A and its associated feature B, their feature interaction is modeled by learning two sets of parameters: 1) the embedding of feature A, and 2) a Multi-Layer Perceptron  ...  There are also some works [18, 20, 26] that exploit meta-path between different nodes for embedding learning.  ... 
arXiv:2011.05625v3 fatcat:rlptqezrqndwtb2kvp23lajbl4

TwHIN: Embedding the Twitter Heterogeneous Information Network for Personalized Recommendation [article]

Ahmed El-Kishky, Thomas Markovich, Serim Park, Chetan Verma, Baekjin Kim, Ramy Eskander, Yury Malkov, Frank Portman, Sofía Samaniego, Ying Xiao, Aria Haghighi
2022 arXiv   pre-print
In this work, we investigate knowledge-graph embeddings for entities in the Twitter HIN (TwHIN); we show that these pretrained representations yield significant offline and online improvement for a diverse  ...  range of downstream recommendation and classification tasks: personalized ads rankings, account follow-recommendation, offensive content detection, and search ranking.  ...  Other approaches have exploited multi-hop meta-paths over HINs to develop collaborative filtering models for personalized recommendation [37] .  ... 
arXiv:2202.05387v1 fatcat:jcjk7kc5bnenfemj2ibysbveze

Learning to Profile: User Meta-Profile Network for Few-Shot Learning [article]

Hao Gong and Qifang Zhao and Tianyu Li and Derek Cho and DuyKhuong Nguyen
2020 arXiv   pre-print
Meta-learning approaches have shown great success in vision and language domains. However, few studies discuss the practice of meta-learning for large-scale industrial applications.  ...  , we propose a meta-learning framework called the Meta-Profile Network, which extends the ideas of matching network and relation network for knowledge transfer and fast adaptation; 2) Encoding strategy  ...  The authors would also like to thank the anonymous reviewers for their valuable comments and helpful suggestions.  ... 
arXiv:2008.12258v2 fatcat:yfawhg6jd5h7dfsse56fnkhcei

Designing Multi-Modal Embedding Fusion-Based Recommender

Anna Wróblewska, Jacek Dąbrowski, Michał Pastuszak, Andrzej Michałowski, Michał Daniluk, Barbara Rychalska, Mikołaj Wieczorek, Sylwia Sysko-Romańczuk
2022 Electronics  
We have developed a machine learning-based recommendation system, which can be easily applied to almost any items and/or actions domain.  ...  We also demonstrate use cases for different e-commerce sites.  ...  Acknowledgments: The authors would like to thank the entire Research Artificial Intelligence team and other Machine Learning engineers from Synerise for joint research experiments in production and substantive  ... 
doi:10.3390/electronics11091391 fatcat:6v6if5e7wjd2tl2ym5roxzz2em

Metapaths guided Neighbors aggregated Network for?Heterogeneous Graph Reasoning [article]

Bang Lin, Xiuchong Wang, Yu Dong, Chengfu Huo, Weijun Ren, Chuanyu Xu
2021 arXiv   pre-print
Heterogeneous graph embedding is to learn the structure and semantic information from the graph, and then embed it into the low-dimensional node representation.  ...  We conduct extensive experiments for the proposed MHN on three real-world heterogeneous graph datasets, including node classification, link prediction and online A/B test on Alibaba mobile application.  ...  Online A/B Test We deploy our inductive model MHN on Alibaba mobile application for it's recall process of recommendation system.  ... 
arXiv:2103.06474v1 fatcat:rt7lwzapebccrg3ta4zrztwc4u

A Survey on Heterogeneous Graph Embedding: Methods, Techniques, Applications and Sources [article]

Xiao Wang and Deyu Bo and Chuan Shi and Shaohua Fan and Yanfang Ye and Philip S. Yu
2020 arXiv   pre-print
We first introduce the basic concepts of HG and discuss the unique challenges brought by the heterogeneity for HG embedding in comparison with homogeneous graph representation learning; and then we systemically  ...  different types of HG embedding methods in the real-world industrial environments for the first time.  ...  HERec [2] aims to learn the embeddings of users and items under different meta-paths and fuses them for recommendation.  ... 
arXiv:2011.14867v1 fatcat:phfoxj7qsrfshfednomeok7pau

User Response Prediction in Online Advertising [article]

Zhabiz Gharibshah, Xingquan Zhu
2021 arXiv   pre-print
Recent years have witnessed a significant increase in the number of studies using computational approaches, including machine learning methods, for user response prediction.  ...  In this survey, we provide a comprehensive review of user response prediction in online advertising and related recommender applications.  ...  Very recently authors in [182] propose to learn transfer component in a cyclic fashion using meta-learning approach for sequential input data.  ... 
arXiv:2101.02342v2 fatcat:clgefamcd5fmbeg5ephizy3zqu

MNI: An enhanced multi-task neighborhood interaction model for recommendation on knowledge graph

Xintao Ma, Liyan Dong, Yuequn Wang, Yongli Li, Hao Zhang, Qi Zhao
2021 PLoS ONE  
Besides, the entities and relations are also semantically embedded.  ...  In this paper, we propose an enhanced multi-task neighborhood interaction (MNI) model for recommendation on knowledge graphs.  ...  Acknowledgments We would like to thank Hao Zhang (Jilin University) for the insightful comments on the manuscript and his guidance and patience enlighten us not only on this paper but also our future.  ... 
doi:10.1371/journal.pone.0258410 pmid:34710122 pmcid:PMC8553089 fatcat:apyneopnl5dujemai4zihownpq

GraphConfRec: A Graph Neural Network-Based Conference Recommender System [article]

Andreea Iana, Heiko Paulheim
2021 arXiv   pre-print
academic careers, or for those seeking to publish outside of their usual domain.  ...  However, choosing a suitable academic venue for publishing one's research can represent a challenging task considering the plethora of available conferences, particularly for those at the start of their  ...  Such dynamic graphs require an inductive embedding method, capable of computing latent representations for new nodes.  ... 
arXiv:2106.12340v1 fatcat:4gwtxtltt5gstlzejvye2pjvhi

Multi-modal Embedding Fusion-based Recommender [article]

Anna Wroblewska
2020 arXiv   pre-print
We have developed a machine learning-based recommendation platform, which can be easily applied to almost any items and/or actions domain.  ...  Recommendation systems have lately been popularized globally, with primary use cases in online interaction systems, with significant focus on e-commerce platforms.  ...  Creation of inductive embeddings (for new nodes) is possible from raw network data using the formula M ′ * Q, where M ′ represents the links between existing and new nodes and Q represents the embeddings  ... 
arXiv:2005.06331v2 fatcat:6nsshrl645codoyzbmsgji6ms4

Multi-Task Feature Learning for Knowledge Graph Enhanced Recommendation [article]

Hongwei Wang, Fuzheng Zhang, Miao Zhao, Wenjie Li, Xing Xie, Minyi Guo
2019 arXiv   pre-print
We propose MKR, a Multi-task feature learning approach for Knowledge graph enhanced Recommendation.  ...  MKR is a deep end-to-end framework that utilizes knowledge graph embedding task to assist recommendation task.  ...  Deep Knowledge-aware Network (DKN) [32] designs a CNN framework to combine entity embeddings with word embeddings for news recommendation.  ... 
arXiv:1901.08907v1 fatcat:ivwxqjp6anamffimhu3sohptwa

Multi-Task Feature Learning for Knowledge Graph Enhanced Recommendation

Hongwei Wang, Fuzheng Zhang, Miao Zhao, Wenjie Li, Xing Xie, Minyi Guo
2019 The World Wide Web Conference on - WWW '19  
We propose MKR, a Multi-task feature learning approach for Knowledge graph enhanced Recommendation.  ...  MKR is a deep end-to-end framework that utilizes knowledge graph embedding task to assist recommendation task.  ...  Deep Knowledge-aware Network (DKN) [32] designs a CNN framework to combine entity embeddings with word embeddings for news recommendation.  ... 
doi:10.1145/3308558.3313411 dblp:conf/www/WangZZLXG19 fatcat:db4ocee6pfcgpb4dxoxxce3id4

Graph Convolutional Embeddings for Recommender Systems

Paula G. Duran, Alexandros Karatzoglou, Jordi Vitria, Xin Xin, Ioannis Arapakis
2021 IEEE Access  
In this work, we generalize the use of GCNs for N-partite graphs by considering N multiple context dimensions and propose a simple way for their seamless integration in modern deep learning RS architectures  ...  More specifically, we define a graph convolutional embedding layer for N-partite graphs that processes user-item-context interactions and constructs node embeddings by leveraging their relational structure  ...  This type of meta-data is commonly known as context [1] . User, item, and context data are often collected in the form of user-id, item-id and context-id's.  ... 
doi:10.1109/access.2021.3096609 fatcat:x4jamauc2fhfnatbw2pbnifkke
« Previous Showing results 1 — 15 out of 3,323 results