TY - CHAP

T1 - Cross-document coreference resolution using latent features

AU - Ngonga Ngomo, Axel Cyrille

AU - Röder, Michael

AU - Usbeck, Ricardo

N1 - European Science Foundation

PY - 2014/10/15

Y1 - 2014/10/15

N2 - Over the last years, entity detection approaches which combine named entity recognition and entity linking have been used to detect mentions of RDF resources from a given reference knowledge base in unstructured data. In this paper, we address the problem of assigning a single URI to named entities which stand for the same real-object across documents but are not yet available in the reference knowledge base. This task is known as cross-document co-reference resolution and has been addressed by manifold approaches in the past. We present a preliminary study of a novel take on the task based on the use of latent features derived from matrix factorizations combined with parameter-free graph clustering. We study the influence of different parameters (window size, rank, hardening) on our approach by comparing the F-measures we achieve on the N3 benchmark. Our results suggest that using latent features leads to higher F-measures with an increase of up to 20.5% on datasets of the N3 collection.

AB - Over the last years, entity detection approaches which combine named entity recognition and entity linking have been used to detect mentions of RDF resources from a given reference knowledge base in unstructured data. In this paper, we address the problem of assigning a single URI to named entities which stand for the same real-object across documents but are not yet available in the reference knowledge base. This task is known as cross-document co-reference resolution and has been addressed by manifold approaches in the past. We present a preliminary study of a novel take on the task based on the use of latent features derived from matrix factorizations combined with parameter-free graph clustering. We study the influence of different parameters (window size, rank, hardening) on our approach by comparing the F-measures we achieve on the N3 benchmark. Our results suggest that using latent features leads to higher F-measures with an increase of up to 20.5% on datasets of the N3 collection.

KW - Informatics

UR - http://www.scopus.com/inward/record.url?scp=84939863962&partnerID=8YFLogxK

M3 - Article in conference proceedings

AN - SCOPUS:84939863962

VL - 1267

T3 - CEUR Workshop Proceedings

SP - 33

EP - 44

BT - Linked Data for Information Extraction 2014.

A2 - Gentile, Anna Lisa

A2 - Zhang, Ziqi

A2 - d'Amato, Claudia

A2 - Paulheim, Heiko

PB - Sun Site Central Europe (RWTH Aachen University)

T2 - 2nd International Workshop on Linked Data for Information Extraction, LD4IE 2014, Co-located with the 13th International Semantic Web Conference, ISWC 2014

Y2 - 20 October 2014

ER -