Logo eprints

Learning Relatedness Measures for Entity Linking

Ceccarelli, Diego and Lucchese, Claudio and Orlando, Salvatore and Perego, Raffaele and Trani, Salvatore Learning Relatedness Measures for Entity Linking. In: Proceedings of the 22nd ACM international conference on Conference on information & knowledge management. ACM, pp. 139-148. ISBN 978-1-4503-2263-8 (2013)

[img]
Preview
PDF
Download (562kB) | Preview

Abstract

Entity Linking is the task of detecting, in text documents, relevant mentions to entities of a given knowledge base. To this end, entity-linking algorithms use several signals and features extracted from the input text or from the knowl- edge base. The most important of such features is entity relatedness. Indeed, we argue that these algorithms benefit from maximizing the relatedness among the relevant enti- ties selected for annotation, since this minimizes errors in disambiguating entity-linking. The definition of an e↵ective relatedness function is thus a crucial point in any entity-linking algorithm. In this paper we address the problem of learning high-quality entity relatedness functions. First, we formalize the problem of learning entity relatedness as a learning-to-rank problem. We propose a methodology to create reference datasets on the basis of manually annotated data. Finally, we show that our machine-learned entity relatedness function performs better than other relatedness functions previously proposed, and, more importantly, improves the overall performance of dif- ferent state-of-the-art entity-linking algorithms.

Item Type: Book Section
Identification Number: 10.1145/2505515.2505711
Uncontrolled Keywords: Entity linking, relatedness measures, learning to rank
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Q Science > QA Mathematics > QA76 Computer software
Research Area: Computer Science and Applications
Depositing User: Users 43 not found.
Date Deposited: 13 Nov 2013 15:00
Last Modified: 28 Jan 2014 15:19
URI: http://eprints.imtlucca.it/id/eprint/1890

Actions (login required)

Edit Item Edit Item