Semantically enhanced entity ranking

verfasst von: Gianluca Demartini, Claudiu S. Firan, Tereza Iofciu, Wolfgang Nejdl
Abstract: Users often want to find entities instead of just documents, i.e., finding documents entirely about specific real-world entities rather than general documents where the entities are merely mentioned. Searching for entities on Web scale repositories is still an open challenge as the effectiveness of ranking is usually not satisfactory. Semantics can be used in this context to improve the results leveraging on entity-driven ontologies. In this paper we propose three categories of algorithms for query adaptation, using (1) semantic information, (2) NLP techniques, and (3) link structure, to rank entities in Wikipedia. Our approaches focus on constructing queries using not only keywords but also additional syntactic information, while semantically relaxing the query relying on a highly accurate ontology. The results show that our approaches perform effectively, and that the combination of simple NLP, Link Analysis and semantic techniques improves the retrieval performance of entity search.
Organisationseinheit(en): Forschungszentrum L3S
Typ: Aufsatz in Konferenzband
Seiten: 176-188
Anzahl der Seiten: 13
Publikationsdatum: 2008
Publikationsstatus: Veröffentlicht
Peer-reviewed: Ja
ASJC Scopus Sachgebiete: Theoretische Informatik, Informatik (insg.)
Elektronische Version(en): https://doi.org/10.1007/978-3-540-85481-4_15 (Zugang: Geschlossen)

BibTeX

@inproceedings{5204a1c0283744409324bf93ea440d15,
title = "Semantically enhanced entity ranking",
abstract = "Users often want to find entities instead of just documents, i.e., finding documents entirely about specific real-world entities rather than general documents where the entities are merely mentioned. Searching for entities on Web scale repositories is still an open challenge as the effectiveness of ranking is usually not satisfactory. Semantics can be used in this context to improve the results leveraging on entity-driven ontologies. In this paper we propose three categories of algorithms for query adaptation, using (1) semantic information, (2) NLP techniques, and (3) link structure, to rank entities in Wikipedia. Our approaches focus on constructing queries using not only keywords but also additional syntactic information, while semantically relaxing the query relying on a highly accurate ontology. The results show that our approaches perform effectively, and that the combination of simple NLP, Link Analysis and semantic techniques improves the retrieval performance of entity search.",
author = "Gianluca Demartini and Firan, {Claudiu S.} and Tereza Iofciu and Wolfgang Nejdl",
year = "2008",
doi = "10.1007/978-3-540-85481-4_15",
language = "English",
isbn = "978-3-540-85480-7",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Berlin Heidelberg",
pages = "176--188",
booktitle = "Web Information Systems Engineering",
note = "9th International Conference on Web Information Systems Engineering, WISE 2008 ; Conference date: 01-09-2008 Through 03-09-2008",
}