In online social networks, most relationships are lack of meaning labels (e.g., “colleague” and “intimate friends”), simply because users do not take the time to label them. An interesting question is: can we automatically infer the type of social relationships in a large network? what are the fundamental factors that imply the type of social relationships? In this work, we formalize the problem of social relationship learning into a semi-supervised framework, and propose a Partially-labeled Pairwise Factor Graph Model (PLP-FGM) for learning to infer the type of social ties. We tested the model on three different genres of data sets: Publication, Email and Mobile. Experimental results demonstrate that the proposed PLP-FGM model can accurately infer 92.7% of advisor-advisee relationships from the coauthor network (Publication), 88.0% of manager-subordinate relationships from the email network (Email), and 83.1% of the friendships from the mobile network (Mobile). Finally, we develop a distributed learning algorithm to scale up the model to real large networks.
Описание
Learning to Infer Social Ties in Large Networks - Springer
%0 Book Section
%1 conf/pkdd/TangZT11
%A Tang, Wenbin
%A Zhuang, Honglei
%A Tang, Jie
%B Proceedings of the ECML/PKDD 2011
%D 2011
%E Gunopulos, Dimitrios
%E Hofmann, Thomas
%E Malerba, Donato
%E Vazirgiannis, Michalis
%I Springer Berlin Heidelberg
%K relationship_mining
%P 381-397
%R 10.1007/978-3-642-23808-6_25
%T Learning to Infer Social Ties in Large Networks
%U http://dblp.uni-trier.de/db/conf/pkdd/pkdd2011-3.html#TangZT11
%V 6913
%X In online social networks, most relationships are lack of meaning labels (e.g., “colleague” and “intimate friends”), simply because users do not take the time to label them. An interesting question is: can we automatically infer the type of social relationships in a large network? what are the fundamental factors that imply the type of social relationships? In this work, we formalize the problem of social relationship learning into a semi-supervised framework, and propose a Partially-labeled Pairwise Factor Graph Model (PLP-FGM) for learning to infer the type of social ties. We tested the model on three different genres of data sets: Publication, Email and Mobile. Experimental results demonstrate that the proposed PLP-FGM model can accurately infer 92.7% of advisor-advisee relationships from the coauthor network (Publication), 88.0% of manager-subordinate relationships from the email network (Email), and 83.1% of the friendships from the mobile network (Mobile). Finally, we develop a distributed learning algorithm to scale up the model to real large networks.
%@ 978-3-642-23807-9
@incollection{conf/pkdd/TangZT11,
abstract = {In online social networks, most relationships are lack of meaning labels (e.g., “colleague” and “intimate friends”), simply because users do not take the time to label them. An interesting question is: can we automatically infer the type of social relationships in a large network? what are the fundamental factors that imply the type of social relationships? In this work, we formalize the problem of social relationship learning into a semi-supervised framework, and propose a Partially-labeled Pairwise Factor Graph Model (PLP-FGM) for learning to infer the type of social ties. We tested the model on three different genres of data sets: Publication, Email and Mobile. Experimental results demonstrate that the proposed PLP-FGM model can accurately infer 92.7% of advisor-advisee relationships from the coauthor network (Publication), 88.0% of manager-subordinate relationships from the email network (Email), and 83.1% of the friendships from the mobile network (Mobile). Finally, we develop a distributed learning algorithm to scale up the model to real large networks.},
added-at = {2015-05-10T10:53:14.000+0200},
author = {Tang, Wenbin and Zhuang, Honglei and Tang, Jie},
biburl = {https://www.bibsonomy.org/bibtex/2857a9becb60a167ce97abccfd6e5e51e/jiangxiluning},
booktitle = {Proceedings of the ECML/PKDD 2011},
description = {Learning to Infer Social Ties in Large Networks - Springer},
doi = {10.1007/978-3-642-23808-6_25},
editor = {Gunopulos, Dimitrios and Hofmann, Thomas and Malerba, Donato and Vazirgiannis, Michalis},
interhash = {10257223148a09d56cd1e3f4e9ea426c},
intrahash = {857a9becb60a167ce97abccfd6e5e51e},
isbn = {978-3-642-23807-9},
keywords = {relationship_mining},
language = {English},
pages = {381-397},
publisher = {Springer Berlin Heidelberg},
series = {Lecture Notes in Computer Science},
timestamp = {2015-05-10T10:53:14.000+0200},
title = {Learning to Infer Social Ties in Large Networks},
type = {Publication},
url = {http://dblp.uni-trier.de/db/conf/pkdd/pkdd2011-3.html#TangZT11},
volume = 6913,
year = 2011
}