Filters








3,352 Hits in 4.8 sec

Learning Attribute-Structure Co-Evolutions in Dynamic Graphs [article]

Daheng Wang, Zhihan Zhang, Yihong Ma, Tong Zhao, Tianwen Jiang, Nitesh V. Chawla, Meng Jiang
2020 arXiv   pre-print
It has a temporal self-attention mechanism to model long-range dependencies in the evolution.  ...  Most graph neural network models learn embeddings of nodes in static attributed graphs for predictive analysis. Recent attempts have been made to learn temporal proximity of the nodes.  ...  It aggregated the information in previous snapshots to the current one using temporal self-attention and employed a multi-task loss function based on attribute inference and link prediction over time.  ... 
arXiv:2007.13004v1 fatcat:2lqeepd4vrgx7nfxwcazjvg7k4

TD score: Time Aware Domain Similarity based Link Prediction

Yugchhaya Galphat
2022 International Journal for Research in Applied Science and Engineering Technology  
Keywords: Co-authorship network; link prediction; node similarity; global feature; temporal feature; domain similarity.  ...  Experiment over co-authorship network reveals that link prediction covering time aware domain similarity is effective and efficient approach than traditional ones.  ...  Temporal graph can be created by encoding temporal data into graphs. This will act as a valuable tool for temporal analysis and prediction.  ... 
doi:10.22214/ijraset.2022.43176 fatcat:qg45ioaconbdjfhq5mvs73irc4

Communities detection and analysis of their dynamics in collaborative networks

Manel Ben Jdidia, Celine Robardet, Eric Fleury
2007 2007 2nd International Conference on Digital Information Management  
We apply this approach on the Infocom co-authorship network to determine stable collaborations and evolving communities.  ...  Finally, we analyse the impact of the co-authorships relation topology on the formation of the program committee board of the conference. # Author cliques C Associated set S of Years 1 M.  ...  To identify evolving communities, we construct the graph G evol on the Infocom co-authorship data. The obtained graph contains 8 692 vertices and 29 972 edges.  ... 
doi:10.1109/icdim.2007.4444313 dblp:conf/icdim/JdidiaRF07 fatcat:xkoprw6yqnaqvexwf7ikh5xlie

Dual network embedding for representing research interests in the link prediction problem on co-authorship networks

Ilya Makarov, Olga Gerasimova, Pavel Sulimov, Leonid E. Zhukov
2019 PeerJ Computer Science  
We present a study on co-authorship network representation based on network embedding together with additional information on topic modeling of research papers and new edge embedding operator.  ...  Standard graph feature engineering and network embedding methods were combined for constructing co-author recommender system formulated as LP problem and prediction of future graph structure.  ...  of graph-based recommender system predicting links in the co-authorship network and incorporating author research interests in collaborative patterns.  ... 
doi:10.7717/peerj-cs.172 pmid:33816825 pmcid:PMC7924522 fatcat:cpddqvz7p5ezvbslcbv54ezrzy

GCN-SE: Attention as Explainability for Node Classification in Dynamic Graphs [article]

Yucai Fan, Yuhang Yao, Carlee Joe-Wong
2021 arXiv   pre-print
less predictive power on the labels.  ...  , e.g., social networks with dynamic relationships, or literature citation networks with changing co-authorships.  ...  In these datasets, every author is extracted as a node in a graph, and co-authorship determines whether there is a connection between two nodes.  ... 
arXiv:2110.05598v1 fatcat:cz555loqifbtbp5rptuu6ont7e

Embedding-Based Recommendations on Scholarly Knowledge Graphs [chapter]

Mojtaba Nayyeri, Sahar Vahdati, Xiaotian Zhou, Hamed Shariat Yazdi, Jens Lehmann
2020 Lecture Notes in Computer Science  
TransE with Soft Margin (TransE-SM) obtains a score of 79.5% Hits@10 for co-authorship link prediction task while the original TransE obtains 77.2%, on the same task.  ...  In this paper, KGEs are reconciled with a specific loss function (Soft Margin) and examined with respect to their performance for co-authorship link prediction task on scholarly KGs.  ...  , -validating the predicted co-authorship links by a profile check of scholars.  ... 
doi:10.1007/978-3-030-49461-2_15 fatcat:37ioh4sql5danl4kc4nppa3oza

Love tHy Neighbour: Remeasuring Local Structural Node Similarity in Hypergraph-Derived Networks [article]

Govind Sharma, Paarth Gupta, M. Narasihma Murty
2021 arXiv   pre-print
Using a combination of the existing graph-based and the proposed hypergraph-based similarity scores as features for a classifier predicts links much better than using the former solely.  ...  We compare our scores with graph-based scores (over clique-expansions of hypergraphs into graphs) from the state-of-the-art.  ...  co-authorship network from its originally occurring hypergraph (i.e., its higher-order counterpart).  ... 
arXiv:2111.00256v1 fatcat:z5unh6xkirdm7bdcwc2na7dlly

Dynamic Graph Representation Learning via Self-Attention Networks [article]

Aravind Sankar, Yanhong Wu, Liang Gou, Wei Zhang, Hao Yang
2019 arXiv   pre-print
We conduct link prediction experiments on two classes of graphs: communication networks and bipartite rating networks.  ...  Previous methods on graph representation learning mainly focus on static graphs, however, many real-world graphs are dynamic and evolve over time.  ...  We leave exploring other architectural design choices based on structural and temporal self-attentions as future work.  ... 
arXiv:1812.09430v2 fatcat:qfilyfvmj5ecjmiknkvoropeqi

ConvDySAT: Deep Neural Representation Learning on Dynamic Graphs via Self-Attention and Convolutional Neural Networks [article]

Ahmad Hafez, Atulya Praphul, Yousef Jaradt, Ezani Godwin
2021 arXiv   pre-print
Learning node representations on temporal graphs is a fundamental step to learn real-word dynamic graphs efficiently.  ...  learning methods focused generally on static graphs.  ...  Secondly our model can be expanded for multi-layer dynamic graph networks and multifeatured graph networks like Co-Authorship dataset.  ... 
arXiv:2106.11430v1 fatcat:t2r2oycql5gttk3z22qblkrk3q

Link Prediction in Social Networks: the State-of-the-Art [article]

Peng Wang and Baowen Xu and Yurong Wu and Xiaoyu Zhou
2014 arXiv   pre-print
A systematical category for link prediction techniques and problems is presented. Then link prediction techniques and problems are analyzed and discussed.  ...  In social networks, link prediction predicts missing links in current networks and new or dissolution links in future networks, is important for mining and analyzing the evolution of social networks.  ...  The co-authorship social network is one of the most popular networks in link prediction. Ichise et al.  ... 
arXiv:1411.5118v2 fatcat:ns5ufekku5hwnotjlfgj2oiaei

Link prediction in social networks: the state-of-the-art

Peng Wang, BaoWen Xu, YuRong Wu, XiaoYu Zhou
2014 Science China Information Sciences  
A systematical category for link prediction techniques and problems is presented. Then link prediction techniques and problems are analyzed and discussed.  ...  In social networks, link prediction predicts missing links in current networks and new or dissolution links in future networks, is important for mining and analyzing the evolution of social networks.  ...  The co-authorship social network is one of the most popular networks in link prediction. Ichise et al.  ... 
doi:10.1007/s11432-014-5237-y fatcat:x6jd4zwg7fgefjrho4en4rtfgm

Tie Strength in Online Social Networks and its Applications: A Brief Study [article]

Chandni Saxena, Tanvir Ahmad
2020 arXiv   pre-print
In online social network (OSN), understanding the factors bound to the role and strength of interaction(tie) are essential to model a wide variety of network-based applications.  ...  The recognition of these interactions can enhance the accuracy of link prediction, improve in ranking of dominants, reliability of recommendation and enhance targeted marketing as decision support system  ...  [6] identified tie strength in temporal network of co-authorship relations by measuring edge persistence and evolution of ties over time.  ... 
arXiv:2002.07608v2 fatcat:2mehz44vhnaovgnrz65o7zodxi

Prediction and ranking algorithms for event-based network data

Joshua O'Madadhain, Jon Hutchins, Padhraic Smyth
2005 SIGKDD Explorations  
In contrast, in this paper we focus on the temporal nature of such data.  ...  Examples include email traffic, telephone calls, and research publications (interpreted as co-authorship events).  ...  Acknowledgements This material is based upon work supported by the National Science Foundation under award number NSF IIS-0083489.  ... 
doi:10.1145/1117454.1117458 fatcat:k4bty55fq5c5zo6xt4bqc7txzi

Temporal Network Representation Learning via Historical Neighborhoods Aggregation [article]

Shixun Huang, Zhifeng Bao, Guoliang Li, Yanghao Zhou, J.Shane Culpepper
2020 arXiv   pre-print
Then we apply a deep learning model which uses a custom attention mechanism to induce node embeddings that directly capture temporal information in the underlying feature representation.  ...  More specifically, we first propose a temporal random walk that can identify relevant nodes in historical neighborhoods which have impact on edge formations.  ...  . • DBLP is a co-author network where each edge represents a co-authorship between two authors (nodes).  ... 
arXiv:2003.13212v1 fatcat:me7k6zbln5cmhnxzoltalskzhu

Temporal Network Representation Learning via Historical Neighborhoods Aggregation

Shixun Huang, Zhifeng Bao, Guoliang Li, Yanghao Zhou, J. Shane Culpepper
2020 2020 IEEE 36th International Conference on Data Engineering (ICDE)  
Then we apply a deep learning model which uses a custom attention mechanism to induce node embeddings that directly capture temporal information in the underlying feature representation.  ...  More specifically, we first propose a temporal random walk that can identify relevant nodes in historical neighborhoods which have impact on edge formations.  ...  . • DBLP is a co-author network where each edge represents a co-authorship between two authors (nodes).  ... 
doi:10.1109/icde48307.2020.00101 dblp:conf/icde/HuangB0ZC20 fatcat:b7nrbfmfrzditgqtznlxvxilxu
« Previous Showing results 1 — 15 out of 3,352 results