Please use this identifier to cite or link to this item: https://hdl.handle.net/1822/39388

TitleTemporal analysis of CHAVE collection
Author(s)Craveiro, Olga
Macedo, Joaquim
Madeira, Henrique
KeywordsTemporal Information Extraction
Temporal Information Retrieval
Issue date5-Sep-2015
PublisherSpringer International Publishing AG
JournalLecture Notes in Computer Science
Abstract(s)The importance of temporal information (TI) is increasing in several Information Retrieval (IR) tasks. CHAVE, available from Linguateca’s site, is the only ad hoc IR test collection with Portuguese texts. So, the research question of this work is whether this collection is sufficiently rich to be used in Temporal IR evaluation. The obtained answer was yes. By the analysis of the CHAVE collection, we verified that 22% of the topics and 86% of the documents have at least one chronon. 49% of topics are time-sensitive. Analyzing the relation of topics with documents, relevant documents of time-sensitive topics converge to a specific date(s), while the non-relevant ones are dispersed along the timeline. Finally, we used a peak dates strategy as a time-aware query expansion (QE) process. Experiments showed effectiveness improvements for time-sensitive queries.
TypeConference paper
DescriptionLecture Notes in Computer Science, 9309
URIhttps://hdl.handle.net/1822/39388
ISBN978-3-319-23825-8
DOI10.1007/978-3-319-23826-5_7
ISSN0302-9743
Publisher versionhttp://link.springer.com/chapter/10.1007/978-3-319-23826-5_7
Peer-Reviewedyes
AccessRestricted access (UMinho)
Appears in Collections:CAlg - Artigos em revistas internacionais / Papers in international journals

Files in This Item:
File Description SizeFormat 
temporal-chave.pdf
  Restricted access
242,47 kBAdobe PDFView/Open

Partilhe no FacebookPartilhe no TwitterPartilhe no DeliciousPartilhe no LinkedInPartilhe no DiggAdicionar ao Google BookmarksPartilhe no MySpacePartilhe no Orkut
Exporte no formato BibTex mendeley Exporte no formato Endnote Adicione ao seu ORCID