The Linking Open Data (LOD) project is an ongoing effort to construct a global data space, i.e. the Web of Data. One important part of this project is to establish owl:sameAs links among structured data sources. Such links indicate equivalent instances that refer to the same real-world object. The problem of discovering owl:sameAs links between pairwise data sources is called instance matching. Most of the existing approaches addressing this problem rely on the quality of prior schema matching, which is not always good enough in the LOD scenario. In this paper, we propose a schema-independent instance-pair similarity metric based on several general descriptive features. We transform the instance matching problem to the binary classification problem and solve it by machine learning algorithms. Furthermore, we employ some transfer learning methods to utilize the existing owl:sameAs links in LOD to reduce the demand for labeled data. We carry out experiments on some datasets of OAEI2010. The results show that our method performs well on real-world LOD data and outperforms the participants of OAEI2010.
Description
A Machine Learning Approach for Instance Matching Based on Similarity Metrics | SpringerLink
%0 Conference Paper
%1 10.1007/978-3-642-35176-1_29
%A Rong, Shu
%A Niu, Xing
%A Xiang, Evan Wei
%A Wang, Haofen
%A Yang, Qiang
%A Yu, Yong
%B The Semantic Web -- ISWC 2012
%C Berlin, Heidelberg
%D 2012
%E Cudré-Mauroux, Philippe
%E Heflin, Jeff
%E Sirin, Evren
%E Tudorache, Tania
%E Euzenat, Jérôme
%E Hauswirth, Manfred
%E Parreira, Josiane Xavier
%E Hendler, Jim
%E Schreiber, Guus
%E Bernstein, Abraham
%E Blomqvist, Eva
%I Springer Berlin Heidelberg
%K instance machinelearning matching
%P 460--475
%T A Machine Learning Approach for Instance Matching Based on Similarity Metrics
%X The Linking Open Data (LOD) project is an ongoing effort to construct a global data space, i.e. the Web of Data. One important part of this project is to establish owl:sameAs links among structured data sources. Such links indicate equivalent instances that refer to the same real-world object. The problem of discovering owl:sameAs links between pairwise data sources is called instance matching. Most of the existing approaches addressing this problem rely on the quality of prior schema matching, which is not always good enough in the LOD scenario. In this paper, we propose a schema-independent instance-pair similarity metric based on several general descriptive features. We transform the instance matching problem to the binary classification problem and solve it by machine learning algorithms. Furthermore, we employ some transfer learning methods to utilize the existing owl:sameAs links in LOD to reduce the demand for labeled data. We carry out experiments on some datasets of OAEI2010. The results show that our method performs well on real-world LOD data and outperforms the participants of OAEI2010.
%@ 978-3-642-35176-1
@inproceedings{10.1007/978-3-642-35176-1_29,
abstract = {The Linking Open Data (LOD) project is an ongoing effort to construct a global data space, i.e. the Web of Data. One important part of this project is to establish owl:sameAs links among structured data sources. Such links indicate equivalent instances that refer to the same real-world object. The problem of discovering owl:sameAs links between pairwise data sources is called instance matching. Most of the existing approaches addressing this problem rely on the quality of prior schema matching, which is not always good enough in the LOD scenario. In this paper, we propose a schema-independent instance-pair similarity metric based on several general descriptive features. We transform the instance matching problem to the binary classification problem and solve it by machine learning algorithms. Furthermore, we employ some transfer learning methods to utilize the existing owl:sameAs links in LOD to reduce the demand for labeled data. We carry out experiments on some datasets of OAEI2010. The results show that our method performs well on real-world LOD data and outperforms the participants of OAEI2010.},
added-at = {2022-05-12T15:44:10.000+0200},
address = {Berlin, Heidelberg},
author = {Rong, Shu and Niu, Xing and Xiang, Evan Wei and Wang, Haofen and Yang, Qiang and Yu, Yong},
biburl = {https://www.bibsonomy.org/bibtex/2d18cd869a516630a96e07aca331b9d6e/simonha94},
booktitle = {The Semantic Web -- ISWC 2012},
description = {A Machine Learning Approach for Instance Matching Based on Similarity Metrics | SpringerLink},
editor = {Cudr{\'e}-Mauroux, Philippe and Heflin, Jeff and Sirin, Evren and Tudorache, Tania and Euzenat, J{\'e}r{\^o}me and Hauswirth, Manfred and Parreira, Josiane Xavier and Hendler, Jim and Schreiber, Guus and Bernstein, Abraham and Blomqvist, Eva},
interhash = {a372f615d0e074fbb923111d6ebf2d67},
intrahash = {d18cd869a516630a96e07aca331b9d6e},
isbn = {978-3-642-35176-1},
keywords = {instance machinelearning matching},
pages = {460--475},
publisher = {Springer Berlin Heidelberg},
timestamp = {2022-05-12T15:44:10.000+0200},
title = {A Machine Learning Approach for Instance Matching Based on Similarity Metrics},
year = 2012
}