Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Continuous-Time Relationship Prediction in Dynamic Heterogeneous Information Networks

Continuous-Time Relationship Prediction in Dynamic Heterogeneous Information Networks Online social networks, World Wide Web, media, and technological networks, and other types of so-called information networks are ubiquitous nowadays. These information networks are inherently heterogeneous and dynamic. They are heterogeneous as they consist of multi-typed objects and relations, and they are dynamic as they are constantly evolving over time. One of the challenging issues in such heterogeneous and dynamic environments is to forecast those relationships in the network that will appear in the future. In this article, we try to solve the problem of continuous-time relationship prediction in dynamic and heterogeneous information networks. This implies predicting the time it takes for a relationship to appear in the future, given its features that have been extracted by considering both heterogeneity and temporal dynamics of the underlying network. To this end, we first introduce a feature extraction framework that combines the power of meta-path-based modeling and recurrent neural networks to effectively extract features suitable for relationship prediction regarding heterogeneity and dynamicity of the networks. Next, we propose a supervised non-parametric approach, called Non-Parametric Generalized Linear Model (Np-Glm), which infers the hidden underlying probability distribution of the relationship building time given its features. We then present a learning algorithm to train Np-Glm and an inference method to answer time-related queries. Extensive experiments conducted on synthetic data and three real-world datasets, namely Delicious, MovieLens, and DBLP, demonstrate the effectiveness of Np-Glm in solving continuous-time relationship prediction problem vis-à-vis competitive baselines. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png ACM Transactions on Knowledge Discovery from Data (TKDD) Association for Computing Machinery

Continuous-Time Relationship Prediction in Dynamic Heterogeneous Information Networks

Download PDF
31 pages

Loading next page...
 
/lp/association-for-computing-machinery/continuous-time-relationship-prediction-in-dynamic-heterogeneous-S8A8wfC9bP

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Association for Computing Machinery
Copyright
Copyright © 2019 ACM
ISSN
1556-4681
eISSN
1556-472X
DOI
10.1145/3333028
Publisher site
See Article on Publisher Site

Abstract

Online social networks, World Wide Web, media, and technological networks, and other types of so-called information networks are ubiquitous nowadays. These information networks are inherently heterogeneous and dynamic. They are heterogeneous as they consist of multi-typed objects and relations, and they are dynamic as they are constantly evolving over time. One of the challenging issues in such heterogeneous and dynamic environments is to forecast those relationships in the network that will appear in the future. In this article, we try to solve the problem of continuous-time relationship prediction in dynamic and heterogeneous information networks. This implies predicting the time it takes for a relationship to appear in the future, given its features that have been extracted by considering both heterogeneity and temporal dynamics of the underlying network. To this end, we first introduce a feature extraction framework that combines the power of meta-path-based modeling and recurrent neural networks to effectively extract features suitable for relationship prediction regarding heterogeneity and dynamicity of the networks. Next, we propose a supervised non-parametric approach, called Non-Parametric Generalized Linear Model (Np-Glm), which infers the hidden underlying probability distribution of the relationship building time given its features. We then present a learning algorithm to train Np-Glm and an inference method to answer time-related queries. Extensive experiments conducted on synthetic data and three real-world datasets, namely Delicious, MovieLens, and DBLP, demonstrate the effectiveness of Np-Glm in solving continuous-time relationship prediction problem vis-à-vis competitive baselines.

Journal

ACM Transactions on Knowledge Discovery from Data (TKDD) – Association for Computing Machinery

Published: Jul 29, 2019

Keywords: Link prediction

References

  • A linear regression model for the analysis of life times
    Aalen, Odd O.
  • On dynamic link inference in heterogeneous networks
    Aggarwal, Cham; Xie, Yan; Yu, Philip S.
  • Tobit models: A survey
    Amemiya, Takeshi
  • Learning deep architectures for AI
    Bengio, Yoshua
  • Analysis of survival data under the proportional hazards model
    Breslow, Norman E.
  • Proceedings of the 2nd workshop on information heterogeneity and fusion in recommender systems (HetRec 2011)
    Cantador, Iván; Brusilovsky, Peter; Kuflik, Tsvi
  • Drug--target interaction prediction by random walk on the heterogeneous network
    Chen, Xing; Liu, Ming-Xi; Yan, Gui-Ying
  • Keras
    Chollet, François
  • Semi-supervised sequence learning
    Dai, Andrew M.; Le, Quoc V.
  • Multi-relational link prediction in heterogeneous information networks
    Davis, Darcy; Lichtenwalter, Ryan; Chawla, Nitesh V.
  • Survey and analysis of temporal link prediction in online social networks
    Dhote, Yugchhaya; Mishra, Nishchol; Sharma, Sanjeev
  • Link prediction and recommendation across heterogeneous social networks
    Dong, Yuxiao; Tang, Jie; Wu, Sen; Tian, Jilei; Chawla, Nitesh V.; Rao, Jinghai; Cao, Huanhuan
  • Temporal link prediction using matrix and tensor factorizations
    Dunlavy, Daniel M.; Kolda, Tamara G.; Acar, Evrim
  • Leveraging network dynamics for improved link prediction
    Hajibagheri, Alireza; Sukthankar, Gita; Lakkaraju, Kiran
  • The movielens datasets: History and context
    Harper, F. Maxwell; Konstan, Joseph A.
  • Evaluating the yield of medical tests
    Jr, Frank E. Harrell; Califf, Robert M.; Pryor, David B.; Lee, Kerry L.; Rosati, Robert A.
  • Long short-term memory
    Hochreiter, Sepp; Schmidhuber, Jürgen
  • Fundamentals of Data Structures
    Horowitz, Ellis; Sahni, Sartaj; Anderson-Freed, Susan
  • Applied Survival Analysis: Regression Modeling of Time-to-Event Data
    Jr, David W. Hosmer; Lemeshow, Stanley; May, Susanne
  • What is Twitter, a social network or a news media? In Proceedings of the 19th International Conference on World Wide Web
    Kwak, Haewoon; Lee, Changhyun; Park, Hosung; Moon, Sue
  • The link prediction problem for social networks
    Liben-Nowell, David; Kleinberg, Jon
  • New perspectives and methods in link prediction
    Lichtenwalter, Ryan N.; Lussier, Jake T.; Chawla, Nitesh V.
  • Link prediction in complex networks: A survey
    Lü, Linyuan; Zhou, Tao
  • Relationship prediction in dynamic heterogeneous information networks
    Fard, Amin Milani; Bagheri, Ebrahim; Wang, Ke
  • A novel time series link prediction method: Learning automata approach
    Moradabadi, Behnaz; Meybodi, Mohammad Reza
  • Cross-temporal link prediction
    Oyama, Satoshi; Hayashi, Kohei; Kashima, Hisashi
  • Temporal link prediction using time series of quasi-local node similarity measures
    Özcan, Alper; Öğüdücü, Şule Gündüz
  • Temporality in link prediction: Understanding social complexity
    Potgieter, Anet; April, Kurt A.; Cooke, Richard J. E.; Osunmakinde, Isaac O.
  • Predicting anchor links between heterogeneous social networks
    Sajadmanesh, S.; Rabiee, H. R.; Khodadadi, A.
  • Temporal link prediction in multi-relational network
    Sett, Niladri; Basu, Saptarshi; Nandi, Sukumar; Singh, Sanasam Ranbir
  • A survey of heterogeneous information network analysis
    Shi, Chuan; Li, Yitong; Zhang, Jiawei; Sun, Yizhou; Philip, S. Yu
  • On ranking in survival analysis: Bounds on the concordance index
    Steck, Harald; Krishnapuram, Balaji; Dehing-Oberije, Cary; Lambin, Philippe; Raykar, Vikas C.
  • Co-author relationship prediction in heterogeneous bibliographic networks
    Sun, Yizhou; Barber, Rick; Gupta, Manish; Aggarwal, Charu C.; Han, Jiawei
  • When will it happen? Relationship prediction in heterogeneous information networks
    Sun, Yizhou; Han, Jiawei; Aggarwal, Charu C.; Chawla, Nitesh V.
  • Pathsim: Meta path-based top-k similarity search in heterogeneous information networks
    Sun, Yizhou; Han, Jiawei; Yan, Xifeng; Yu, Philip S.; Wu, Tianyi
  • ArnetMiner: Extraction and mining of academic social networks
    Tang, Jie; Zhang, Jing; Yao, Limin; Li, Juanzi; Zhang, Li; Su, Zhong
  • Link prediction in relational data
    Taskar, Ben; Wong, Ming-Fai; Abbeel, Pieter; Koller, Daphne
  • Dynamic modularity in protein interaction networks predicts breast cancer outcome
    Taylor, Ian W.; Linding, Rune; Warde-Farley, David; Liu, Yongmei; Pesquita, Catia; Faria, Daniel; Bull, Shelley; Pawson, Tony; Morris, Quaid; Wrana, Jeffrey L
  • Estimation of relationships for limited dependent variables
    Tobin, James
  • Local probabilistic models for link prediction
    Wang, Chao; Satuluri, Venu; Parthasarathy, Srinivasan
  • Link prediction in social networks: The state-of-the-art
    Wang, Peng; Xu, BaoWen; Wu, YuRong; Zhou, XiaoYu
  • A review of link prediction in social networks
    Wang, Tingli; Liao, Guoqiong
  • Social Network Analysis: Methods and Applications
    Wasserman, Stanley; Faust, Katherine
  • Temporally factorized network modeling for evolutionary network analysis
    Yu, Wenchao; Aggarwal, Charu C; Wang, Wei
  • ADADELTA: An adaptive learning rate method
    Zeiler, Matthew D.
  • Meta-path based multi-network collective link prediction
    Zhang, Jiawei; Yu, Philip S.; Zhou, Zhi-Hua