V. Leroy, B. Cambazoglu, and F. Bonchi. Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, page 393--402. New York, NY, USA, ACM, (2010)
DOI: 10.1145/1835804.1835855
Abstract
In the traditional link prediction problem, a snapshot of a social network is used as a starting point to predict, by means of graph-theoretic measures, the links that are likely to appear in the future. In this paper, we introduce cold start link prediction as the problem of predicting the structure of a social network when the network itself is totally missing while some other information regarding the nodes is available. We propose a two-phase method based on the bootstrap probabilistic graph. The first phase generates an implicit social network under the form of a probabilistic graph. The second phase applies probabilistic graph-based measures to produce the final prediction. We assess our method empirically over a large data collection obtained from Flickr, using interest groups as the initial information. The experiments confirm the effectiveness of our approach.
%0 Conference Paper
%1 Leroy:2010:CSL:1835804.1835855
%A Leroy, Vincent
%A Cambazoglu, B. Barla
%A Bonchi, Francesco
%B Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
%C New York, NY, USA
%D 2010
%I ACM
%K thema:exploiting_place_features_in_link_prediction_on_location-based_social_networks
%P 393--402
%R 10.1145/1835804.1835855
%T Cold Start Link Prediction
%U http://doi.acm.org/10.1145/1835804.1835855
%X In the traditional link prediction problem, a snapshot of a social network is used as a starting point to predict, by means of graph-theoretic measures, the links that are likely to appear in the future. In this paper, we introduce cold start link prediction as the problem of predicting the structure of a social network when the network itself is totally missing while some other information regarding the nodes is available. We propose a two-phase method based on the bootstrap probabilistic graph. The first phase generates an implicit social network under the form of a probabilistic graph. The second phase applies probabilistic graph-based measures to produce the final prediction. We assess our method empirically over a large data collection obtained from Flickr, using interest groups as the initial information. The experiments confirm the effectiveness of our approach.
%@ 978-1-4503-0055-1
@inproceedings{Leroy:2010:CSL:1835804.1835855,
abstract = {In the traditional link prediction problem, a snapshot of a social network is used as a starting point to predict, by means of graph-theoretic measures, the links that are likely to appear in the future. In this paper, we introduce cold start link prediction as the problem of predicting the structure of a social network when the network itself is totally missing while some other information regarding the nodes is available. We propose a two-phase method based on the bootstrap probabilistic graph. The first phase generates an implicit social network under the form of a probabilistic graph. The second phase applies probabilistic graph-based measures to produce the final prediction. We assess our method empirically over a large data collection obtained from Flickr, using interest groups as the initial information. The experiments confirm the effectiveness of our approach.},
acmid = {1835855},
added-at = {2014-04-24T15:04:54.000+0200},
address = {New York, NY, USA},
author = {Leroy, Vincent and Cambazoglu, B. Barla and Bonchi, Francesco},
biburl = {https://www.bibsonomy.org/bibtex/2075ddef91e76dac87ef6efe1547e15ce/b.helmerich},
booktitle = {Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining},
description = {Cold start link prediction},
doi = {10.1145/1835804.1835855},
interhash = {660d904a7cdd03a447d18dce03316c06},
intrahash = {075ddef91e76dac87ef6efe1547e15ce},
isbn = {978-1-4503-0055-1},
keywords = {thema:exploiting_place_features_in_link_prediction_on_location-based_social_networks},
location = {Washington, DC, USA},
numpages = {10},
pages = {393--402},
publisher = {ACM},
series = {KDD '10},
timestamp = {2014-04-24T15:04:54.000+0200},
title = {Cold Start Link Prediction},
url = {http://doi.acm.org/10.1145/1835804.1835855},
year = 2010
}