Using Named Entity Recognition to Identify Personification Constructions in an English <> Spanish Intermodal Corpus of the EP Committee on Petitions

Conference

Fernando Sánchez Rodas
Simposio Doctoral sobre Procesamiento del Lenguaje Natural, Red temática PLN.net, Baeza (Spain), 2021 Oct 19

PDF

Cite

APA Click to copy
Rodas, F. S. (2021). Using Named Entity Recognition to Identify Personification Constructions in an English <> Spanish Intermodal Corpus of the EP Committee on Petitions. In Simposio Doctoral sobre Procesamiento del Lenguaje Natural. Baeza (Spain): Red temática PLN.net.

Chicago/Turabian Click to copy
Rodas, Fernando Sánchez. “Using Named Entity Recognition to Identify Personification Constructions in an English ≪≫ Spanish Intermodal Corpus of the EP Committee on Petitions.” In Simposio Doctoral Sobre Procesamiento Del Lenguaje Natural. Baeza (Spain): Red temática PLN.net, 2021.

MLA Click to copy
Rodas, Fernando Sánchez. “Using Named Entity Recognition to Identify Personification Constructions in an English ≪≫ Spanish Intermodal Corpus of the EP Committee on Petitions.” Simposio Doctoral Sobre Procesamiento Del Lenguaje Natural, Red temática PLN.net, 2021.

BibTeX Click to copy

@conference{fernando2021a,
  title = {Using Named Entity Recognition to Identify Personification Constructions in an English <> Spanish Intermodal Corpus of the EP Committee on Petitions},
  year = {2021},
  month = oct,
  day = {19},
  address = {Baeza (Spain)},
  organization = {Red temática PLN.net},
  author = {Rodas, Fernando Sánchez},
  booktitle = {Simposio Doctoral sobre Procesamiento del Lenguaje Natural},
  month_numeric = {10}
}

Abstract
This paper presents a PhD project which tests the effectiveness of NLP-based methods to extract and analyze large amounts of data from translation and interpreting corpora. More specifically, Named Entity Recognition (NER) applications are used in combination with an intermodal corpus of EU texts (that is, a multilingual corpus which contain all possible variations of mediated and non-mediated discourse) in order to identify personification constructions, especially those related to organizations. From the point of view of Construction Grammar (CxG), personifications are argument-structure constructions loaded with relational meaning, which makes them valuable data to feed Machine Translation (MT) and Machine Interpreting (MI) systems or related electronic tools used by translation and interpreting professionals in the briefing preparation phase. In the future, we expect that the compiled intermodal corpus (named PETIMOD) and the NLP techniques can be used to study further types of constructions in institutional discourse, which would be an important contribution not only to corpus-based translation and interpreting studies, but also to CxG itself.