Error analysis for the improvement of subject ellipsis detection

This paper presents an analysis of the errors of a machine learning method that allow us to propose changes to improve it in future developments. The evaluated system detects Spanish subject ellipsis and yields an accuracy of 85.3%. We extract the wrongly classified instances of our training data (1...

Descripción completa

Detalles Bibliográficos
Autores: Rello, Luz, 1984-, Ferraro, Gabriela, Burga Díaz, Alicia
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2011
País:España
Institución:Universitat Pompeu Fabra
Repositorio:Repositorio Digital de la UPF
OAI Identifier:oai:repositori.upf.edu:10230/43843
Acceso en línea:http://hdl.handle.net/10230/43843
Access Level:acceso abierto
Palabra clave:Subject ellipsis
Impersonal construction
Zero pronoun
Error analysis
Linguistic analysis
Machine learning
Elipsis de sujeto
Construcción impersonal
Pronombre zero
Análisis de errores
Análisis lingüístico
Aprendizaje automático
Descripción
Sumario:This paper presents an analysis of the errors of a machine learning method that allow us to propose changes to improve it in future developments. The evaluated system detects Spanish subject ellipsis and yields an accuracy of 85.3%. We extract the wrongly classified instances of our training data (1,001) and classify the errors. We perform an analysis of these instances taking into account the features and the linguistic patterns involved, which motivate the inclusion of new features and rules in the system.