FEMsum: A flexible eclectic multitask summarizer architecture evaluated in multidocument tasks

This article describes two types of summarization approaches integrated in a flexible architecture for multitask summarization. The first type is based on the use of lexical features, while the second one is grounded on syntactic and semantic information. All the approaches have been evaluated in ex...

Descripción completa

Detalles Bibliográficos
Autores: Fuentes Fort, Maria, Rodríguez Hontoria, Horacio|||0000-0002-5314-6631, Turmo Borras, Jorge|||0000-0002-7521-1115
Tipo de recurso: informe técnico
Fecha de publicación:2007
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/87422
Acceso en línea:https://hdl.handle.net/2117/87422
Access Level:acceso abierto
Palabra clave:Text Summarization
Spontaneous Speech Summarization
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Llenguatge natural
Descripción
Sumario:This article describes two types of summarization approaches integrated in a flexible architecture for multitask summarization. The first type is based on the use of lexical features, while the second one is grounded on syntactic and semantic information. All the approaches have been evaluated in experiments where, given a set of documents, they are expected to produce summaries answering a user need (expressed by a query) in a reduced set of relevant textual fragments. Their performance is analyzed in two different tasks: written news and scientific oral presentations.