We propose a graph-based semantic model for representing document content. Our method relies on the use of a semantic network, namely the DBpedia knowledge base, for acquiring fine-grained information about entities and their semantic relations, thus resulting in a knowledge-rich document model. We demonstrate the benefits of these semantic representations in two tasks: entity ranking and computing document semantic similarity. To this end, we couple DBpedia's structure with an information-theoretic measure of concept association, based on its explicit semantic relations, and compute semantic similarity using a Graph Edit Distance based measure, which finds the optimal matching between the documents' entities using the Hungarian method. Experimental results show that our general model outperforms baselines built on top of traditional methods, and achieves a performance close to that of highly specialized methods that have been tuned to these specific tasks.
%0 Conference Paper
%1 Schuhmacher:2014:KGD:2556195.2556250
%A Schuhmacher, Michael
%A Ponzetto, Simone Paolo
%B Proceedings of the 7th ACM International Conference on Web Search and Data Mining
%C New York, NY, USA
%D 2014
%I ACM
%K document graph model semantic structured
%P 543--552
%R 10.1145/2556195.2556250
%T Knowledge-based Graph Document Modeling
%U http://doi.acm.org/10.1145/2556195.2556250
%X We propose a graph-based semantic model for representing document content. Our method relies on the use of a semantic network, namely the DBpedia knowledge base, for acquiring fine-grained information about entities and their semantic relations, thus resulting in a knowledge-rich document model. We demonstrate the benefits of these semantic representations in two tasks: entity ranking and computing document semantic similarity. To this end, we couple DBpedia's structure with an information-theoretic measure of concept association, based on its explicit semantic relations, and compute semantic similarity using a Graph Edit Distance based measure, which finds the optimal matching between the documents' entities using the Hungarian method. Experimental results show that our general model outperforms baselines built on top of traditional methods, and achieves a performance close to that of highly specialized methods that have been tuned to these specific tasks.
%@ 978-1-4503-2351-2
@inproceedings{Schuhmacher:2014:KGD:2556195.2556250,
abstract = {We propose a graph-based semantic model for representing document content. Our method relies on the use of a semantic network, namely the DBpedia knowledge base, for acquiring fine-grained information about entities and their semantic relations, thus resulting in a knowledge-rich document model. We demonstrate the benefits of these semantic representations in two tasks: entity ranking and computing document semantic similarity. To this end, we couple DBpedia's structure with an information-theoretic measure of concept association, based on its explicit semantic relations, and compute semantic similarity using a Graph Edit Distance based measure, which finds the optimal matching between the documents' entities using the Hungarian method. Experimental results show that our general model outperforms baselines built on top of traditional methods, and achieves a performance close to that of highly specialized methods that have been tuned to these specific tasks.},
acmid = {2556250},
added-at = {2014-10-24T19:38:13.000+0200},
address = {New York, NY, USA},
author = {Schuhmacher, Michael and Ponzetto, Simone Paolo},
biburl = {https://www.bibsonomy.org/bibtex/2c34580fb709a8fcafa0d1a17e61fa760/jil},
booktitle = {Proceedings of the 7th ACM International Conference on Web Search and Data Mining},
doi = {10.1145/2556195.2556250},
interhash = {659f81fe77ad2fa138c29dd507eeeba9},
intrahash = {c34580fb709a8fcafa0d1a17e61fa760},
isbn = {978-1-4503-2351-2},
keywords = {document graph model semantic structured},
location = {New York, New York, USA},
numpages = {10},
pages = {543--552},
publisher = {ACM},
series = {WSDM '14},
timestamp = {2014-10-24T19:38:13.000+0200},
title = {Knowledge-based Graph Document Modeling},
url = {http://doi.acm.org/10.1145/2556195.2556250},
year = 2014
}