EASSIGNER: DESIGN AND MODELING SOFTWARE FOR AUTOMATING PHILOLOGICAL-LINGUISTIC ANNOTATIONS IN ELECT

Abstract

ABSTRACT

This research work intends to discuss and help solve the problem of inconsistencies generated by the lack of efficiency of the manual philological editing process that can cause noise in the linguistic corpus annotation, which may compromise its morphosyntactic labeling and syntactic annotation, causing a loss of linguistic research. To solve this problem, this research discusses the importance of the use of Information Technology as a tool to help the work of linguists, particularly with regard to Corpus Linguistics. Thus, throughout this work, we define the terms Corpus Linguistics, Computational Linguistics and current uses of Artificial Intelligence are described emphasizing the problems related to Linguistics and the modern strategies in search of solutions. From there, the research describes the importance of the use of annotation softwares in electronic corpora and the consequent need to automate some of these operations through the project for software development. Thus, the design and modeling of the eAssigner tool was also made, presenting its characteristics, limitations and stage of development. To illustrate the need to use the software are presented results of some tests performed on Corpus DoViC document samples.