Filters








13,635 Hits in 4.8 sec

Heterogeneous Edge Embeddings for Friend Recommendation [article]

Janu Verma, Srishti Gupta, Debdoot Mukherjee, Tanmoy Chakraborty
2019 arXiv   pre-print
We propose a friend recommendation system (an application of link prediction) using edge embeddings on social networks.  ...  We evaluate our model on a real-world, active social network where this system is deployed for friend recommendation for millions of users.  ...  The system has two components -network embedding for a large heterogeneous network, and training a friend recommendation model on a large set of known friend and non-friend pairs by leveraging the learned  ... 
arXiv:1902.03124v1 fatcat:pp56atx43zfmbmmv4mdngvwy7q

Exploring Student Check-In Behavior for Improved Point-of-Interest Prediction

Mengyue Hang, Ian Pytlarz, Jennifer Neville
2018 Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining - KDD '18  
We also show how our learned embeddings could be used to identify similar students (e.g., for friend suggestions).  ...  We propose a heterogeneous graph-based method to encode the correlations between users, POIs, and activities, and then jointly learn embeddings for the vertices.  ...  Government is authorized to reproduce and distribute reprints for governmental purposes notwithstanding any copyright notation hereon.  ... 
doi:10.1145/3219819.3219902 dblp:conf/kdd/HangPN18 fatcat:otcygqdcerffhlzagaxgtm23qe

Tripartite Heterogeneous Graph Propagation for Large-scale Social Recommendation [article]

Kyung-Min Kim, Donghyun Kwak, Hanock Kwak, Young-Jin Park, Sangkwon Sim, Jae-Han Cho, Minkyu Kim, Jihun Kwon, Nako Sung, Jung-Woo Ha
2019 arXiv   pre-print
However, various challenging issues of social graphs hinder the practical usage of GNNs for social recommendation, such as their complex noisy connections and high heterogeneity.  ...  The oversmoothing of GNNs is an obstacle of GNN-based social recommendation as well. Here we propose a new graph embedding method Heterogeneous Graph Propagation (HGP) to tackle these issues.  ...  ACKNOWLEDGMENTS The authors appreciate Andy Ko for insightful comments and discussion.  ... 
arXiv:1908.02569v1 fatcat:br7cbjrybzat7o7sr2zuvryf5y

SgWalk: Location Recommendation by User Subgraph Based Graph Embedding

Deniz Canturk, Pinar Karagoz
2021 IEEE Access  
For example, an LBSN graph with user (U), location (L), and friend (F) entities contains not only the homogeneous edges between users (friendship) but also heterogeneous edges linking a user and location  ...  [20] is Meta-path Aggregated Graph Neural Network for heterogeneous graph embedding.  ... 
doi:10.1109/access.2021.3116226 fatcat:s64aypfir5bd3hkrfanu2xztry

Social Recommendation in Heterogeneous Evolving Relation Network [chapter]

Bo Jiang, Zhigang Lu, Yuling Liu, Ning Li, Zelin Cui
2020 Lecture Notes in Computer Science  
One of the main solutions proposed for this information overload problem are recommender systems, which provide personalized results.  ...  The learned evolving relation network is a heterogeneous information network, where the strength of relation between users is a sum of the influence of all historical events.  ...  Specifically, on the one hand, users often have different preferences for different items. On the other hand, user are likely to accept their friends' recommendations.  ... 
doi:10.1007/978-3-030-50371-0_41 fatcat:ih2hi47tkreehe4osodb2ana7i

An Effective Two-way Metapath Encoder over Heterogeneous Information Network for Recommendation

Yanbin Jiang, Huifang Ma, Xiaohui Zhang, Zhixin Li, Liang Chang
2022 Proceedings of the 2022 International Conference on Multimedia Retrieval  
To tackle these limitations, we propose a novel recommendation model with two-way metapath encoder for top-N recommendation, which models metapath similarity and sequence relation dependency in HIN to  ...  Heterogeneous information networks (HINs) are widely used in recommender system research due to their ability to model complex auxiliary information beyond historical interactions to alleviate data sparsity  ...  HINs have been proposed as a general method for learning heterogeneous graphs due to their advantages in modeling heterogeneous graphs containing multiple types of nodes and edges [14] .  ... 
doi:10.1145/3512527.3531402 fatcat:zn7o7j6xyrb7zdkjormf5nw434

Heterogeneous Social Recommendation Model with Network Embedding

Chang Su, Zongchao Hu, Xianzhong Xie
2020 IEEE Access  
social relations on the ranking list; 2) We propose a heterogeneous social recommendation model with network embedding, which effectively analyze the social characteristics among users for recommendation  ...  CONCLUSION In this paper, we propose a heterogeneous social recommendation model with network embedding to effectively integrate social relations into the recommendation model.  ... 
doi:10.1109/access.2020.3038022 fatcat:xcapp4isljeanbjyqp6vwqrms4

Graph Learning Augmented Heterogeneous Graph Neural Network for Social Recommendation [article]

Yiming Zhang, Lingfei Wu, Qi Shen, Yitong Pang, Zhihua Wei, Fangli Xu, Ethan Chang, Bo Long
2021 arXiv   pre-print
In this work, we propose an end-to-end heterogeneous global graph learning framework, namely Graph Learning Augmented Heterogeneous Graph Neural Network (GL-HGNN) for social recommendation.  ...  Since social network (user-user relations) and user-item interactions are both naturally represented as graph-structured data, Graph Neural Networks (GNNs) have thus been widely applied for social recommendation  ...  In order to address these issues, we propose an end-to-end heterogeneous global graph learning framework, namely Graph Learning Augmented Heterogeneous Graph Neural Network (GL-HGNN) for social recommendation  ... 
arXiv:2109.11898v1 fatcat:4ru4bcrgrnh4jf4awn57wsic4q

UniWalk: Explainable and Accurate Recommendation for Rating and Network Data [article]

Haekyu Park, Hyunsik Jeon, Junghwan Kim, Beunguk Ahn, U Kang
2017 arXiv   pre-print
Exploiting both ratings and social graph for recommendation, however, is not trivial because of the heterogeneity of the data.  ...  How can we leverage social network data and observed ratings to correctly recommend proper items and provide a persuasive explanation for the recommendations?  ...  UniWalk provides the best explainability and accuracy for recommendation. Future works include extending the method for distributed systems for scalable learning.  ... 
arXiv:1710.07134v1 fatcat:rscrj44ytbb4nbz4usprdom4sq

H2Rec: Homogeneous and Heterogeneous Network Embedding Fusion for Social Recommendation

Yabin Shao, Cheng Liu
2021 International Journal of Computational Intelligence Systems  
To address these issues, we propose a unified H2Rec model to fuse homogeneous and heterogeneous information for recommendations in social networks.  ...  At present, most social recommendation is based on the homogeneity or heterogeneity of social networks.  ...  For any two users u i ∈ and u j ∈ , if user u i has a social relationship with user u j (such as a trust relationship or a friend relationship), there will be an edge e ij ∈ uu from u i to u j ; otherwise  ... 
doi:10.2991/ijcis.d.210406.001 fatcat:3tvxhnn4vffmzffvegnbskc7uy

Recent Advances in Heterogeneous Relation Learning for Recommendation [article]

Chao Huang
2021 arXiv   pre-print
Finally, we present an exploratory outlook to highlight several promising directions and opportunities in heterogeneous relational learning frameworks for recommendation.  ...  We discuss the learning approaches in each category, such as matrix factorization, attention mechanism and graph neural networks, for effectively distilling heterogeneous contextual information.  ...  Figure 1 : 1 Illustration of relation heterogeneity learning for recommender systems.  ... 
arXiv:2110.03455v1 fatcat:fskj4qdsibfnxefklazdli3tgu

LBSN2Vec++: Heterogeneous Hypergraph Embedding for Location-Based Social Networks

Dingqi Yang, Bingqing Qu, Jie Yang, Philippe Cudre-Mauroux
2020 IEEE Transactions on Knowledge and Data Engineering  
Against this background, we propose in this paper LBSN2Vec++, a heterogeneous hypergraph embedding approach designed specifically for LBSN data for automatic feature learning.  ...  Specifically, LBSN data intrinsically forms a heterogeneous hypergraph including both user-user homogeneous edges (friendships) and user-time-POI-semantic heterogeneous hyperedges (check-ins).  ...  If we learn user and POI embeddings in a unique vector space, this user should be closely surrounded by her friends in the embedding vector space because of friendship edges, and this user should also  ... 
doi:10.1109/tkde.2020.2997869 fatcat:m5zamr37rzectgo5lyciuvupry

Connecting Latent ReLationships over Heterogeneous Attributed Network for Recommendation [article]

Ziheng Duan, Yueyang Wang, Weihao Ye, Zixuan Feng, Qilin Fan, Xiuhua Li
2021 arXiv   pre-print
Graph Neural Network (GNN), which can generate high-quality embeddings by capturing graph-structured information, is convenient for the recommendation.  ...  They cannot characterize heterogeneous and complex data in the recommendation system.  ...  GATNE's embedding of edge heterogeneity and GraphRec's social aggregation cannot be fully utilized in this case.  ... 
arXiv:2103.05749v1 fatcat:xii4gjvm6ren5e2rxroajmxr74

Group-Buying Recommendation for Social E-Commerce [article]

Jun Zhang, Chen Gao, Depeng Jin, Yong Li
2020 arXiv   pre-print
Group-buying recommendation for social e-commerce, which recommends an item list when users want to launch a group, plays an important role in the group success ratio and sales.  ...  However, designing a personalized recommendation model for group buying is an entirely new problem that is seldom explored.  ...  For directed edges in graphs, considering a group-buying behavior b = m i , n, M p , where M p = {m p1 , m p2 , · · · , m p |Mp | }, G i contains a bidirectional edge (m i , n), G p contains |M p | bidirectional  ... 
arXiv:2010.06848v2 fatcat:ewdu4gbz7zed3pwfzd3mybqwee

Genetic Meta-Structure Search for Recommendation on Heterogeneous Information Network

Zhenyu Han, Fengli Xu, Jinghan Shi, Yu Shang, Haorui Ma, Pan Hui, Yong Li
2020 Proceedings of the 29th ACM International Conference on Information & Knowledge Management  
In the past decade, the heterogeneous information network (HIN) has become an important methodology for modern recommender systems.  ...  To address these challenges, we propose Genetic Meta-Structure Search (GEMS) to automatically optimize meta-structure designs for recommendation on HINs.  ...  Heterogeneous nodes and edges make it more difficult to filter out useful information in HINs.  ... 
doi:10.1145/3340531.3412015 dblp:conf/cikm/HanXSSMHL20 fatcat:25nh7nhjyrcyta5ufu5im2d57m
« Previous Showing results 1 — 15 out of 13,635 results