H. Wang, and W. Li. Proceedings of the Twenty-Third International Joint Conference on Artificial Intelligence, page 2726--2732. AAAI Press, (2013)
Abstract
With the emergence of large-scale evolving (time-varying) networks, dynamic network analysis (DNA) has become a very hot research topic in recent years. Although a lot of DNA methods have been proposed by researchers from different communities, most of them can only model snapshot data recorded at a very rough temporal granularity. Recently, some models have been proposed for DNA which can be used to model large-scale citation networks at a fine temporal granularity. However, they suffer from a significant decrease of accuracy over time because the learned parameters or node features are static (fixed) during the prediction process for evolving citation networks. In this paper, we propose a novel model, called online egocentric model (OEM), to learn time-varying parameters and node features for evolving citation networks. Experimental results on real-world citation networks show that our OEM can not only prevent the prediction accuracy from decreasing over time but also uncover the evolution of topics in citation networks.
%0 Conference Paper
%1 wang2013online
%A Wang, Hao
%A Li, Wu-Jun
%B Proceedings of the Twenty-Third International Joint Conference on Artificial Intelligence
%D 2013
%I AAAI Press
%K analysis citation dynamic network prediction sota
%P 2726--2732
%T Online Egocentric Models for Citation Networks
%U http://dl.acm.org/citation.cfm?id=2540128.2540521
%X With the emergence of large-scale evolving (time-varying) networks, dynamic network analysis (DNA) has become a very hot research topic in recent years. Although a lot of DNA methods have been proposed by researchers from different communities, most of them can only model snapshot data recorded at a very rough temporal granularity. Recently, some models have been proposed for DNA which can be used to model large-scale citation networks at a fine temporal granularity. However, they suffer from a significant decrease of accuracy over time because the learned parameters or node features are static (fixed) during the prediction process for evolving citation networks. In this paper, we propose a novel model, called online egocentric model (OEM), to learn time-varying parameters and node features for evolving citation networks. Experimental results on real-world citation networks show that our OEM can not only prevent the prediction accuracy from decreasing over time but also uncover the evolution of topics in citation networks.
%@ 978-1-57735-633-2
@inproceedings{wang2013online,
abstract = {With the emergence of large-scale evolving (time-varying) networks, dynamic network analysis (DNA) has become a very hot research topic in recent years. Although a lot of DNA methods have been proposed by researchers from different communities, most of them can only model snapshot data recorded at a very rough temporal granularity. Recently, some models have been proposed for DNA which can be used to model large-scale citation networks at a fine temporal granularity. However, they suffer from a significant decrease of accuracy over time because the learned parameters or node features are static (fixed) during the prediction process for evolving citation networks. In this paper, we propose a novel model, called online egocentric model (OEM), to learn time-varying parameters and node features for evolving citation networks. Experimental results on real-world citation networks show that our OEM can not only prevent the prediction accuracy from decreasing over time but also uncover the evolution of topics in citation networks.},
acmid = {2540521},
added-at = {2014-01-14T09:28:55.000+0100},
author = {Wang, Hao and Li, Wu-Jun},
biburl = {https://www.bibsonomy.org/bibtex/2cdb3dc8104c39703a1c4a082bf293df7/jaeschke},
booktitle = {Proceedings of the Twenty-Third International Joint Conference on Artificial Intelligence},
interhash = {04d7a9ee89fc483fc1b202000d624944},
intrahash = {cdb3dc8104c39703a1c4a082bf293df7},
isbn = {978-1-57735-633-2},
keywords = {analysis citation dynamic network prediction sota},
location = {Beijing, China},
numpages = {7},
pages = {2726--2732},
publisher = {AAAI Press},
series = {IJCAI'13},
timestamp = {2014-07-28T15:57:31.000+0200},
title = {Online Egocentric Models for Citation Networks},
url = {http://dl.acm.org/citation.cfm?id=2540128.2540521},
year = 2013
}