Romanian TreeAnnotator Cover Image

Romanian TreeAnnotator
Romanian TreeAnnotator

Author(s): Daniela GÎFU
Subject(s): Social Sciences, Education, Higher Education
Published by: Carol I National Defence University Publishing House
Keywords: TreeAnnotator, political corpus; written language; syntactic annotation; NLP;

Summary/Abstract: E-learning is the most popular and effective study method centered on any kind of receptor eager to acquire new knowledge. Due the difference between generations, including the politicians, the language becomes a dynamic organism, meaningful in a social context. It is known that political discourse is influenced by the historical destiny of the society to which is addressed. It copies or it battles with the discursive manner of the great political figures of a certain people. It incorporates the cultural heritage of the listeners as well as the orator's intellectual amplitude. This paper describes a diachronic TreeAnnotator, which is a very important linguistic resource in different comparative syntactic levels. It includes a wide range of data (e.g. text corpora belonging to different registers of language) that can be used for training, testing and evaluation of a parser for Romanian. The aim of this study is to create a starting point for joint Romanian language annotations in order to allow quick and easy access to the resources needed for processing political language. This tool considers a political corpus (Ion C. Bratianu, 1857, one of the famous Romanian politicians and prime-minister) annotated at the syntactic level. The concept behind this method is that the manner of adapting political discourse to the social, cultural, economic or religious context of the public is of paramount importance. This endeavour is meaningful and useful for linguists, specialists in political sciences, computer scientists, being helpful to non-native Romanians who want to study the different linguistic phenomena of our language.

  • Issue Year: 15/2019
  • Issue No: 01
  • Page Range: 181-187
  • Page Count: 7
  • Language: English