SELEcTor: Discovering Similar Entities on LinkEd DaTa by Ranking Their Features

Several approaches have been used in the last years to compute similarity between entities. In this paper, we present a novel approach to compute similarity between entities using their features available as Linked Data. The key idea of the proposed framework, called SELEcTor, is to exploit ranked l...

Full description

Saved in:
Bibliographic Details
Published in2017 IEEE 11th International Conference on Semantic Computing (ICSC) pp. 117 - 124
Main Authors Ruback, Livia, Casanova, Marco Antonio, Renso, Chiara, Lucchese, Claudio
Format Conference Proceeding
LanguageEnglish
Published IEEE 2017
Subjects
Correlation
Data mining
Entity Similarity
Feature extraction
Linked Data
Measurement
Rank correlation
Semantics
Online AccessGet full text

Cover

More Information
Summary:Several approaches have been used in the last years to compute similarity between entities. In this paper, we present a novel approach to compute similarity between entities using their features available as Linked Data. The key idea of the proposed framework, called SELEcTor, is to exploit ranked lists of features extracted from Linked Data sources as a representation of the entities we want to compare. The similarity between two entities is thus mapped to the problem of comparing two ranked lists. Our experiments, conducted with museum data from DBpedia, demonstrate that SELEcTor achieves better accuracy than stateof-the-art methods.
DOI:10.1109/ICSC.2017.46