Model for automatic detection of lexical-syntactic errors in texts written in Spanish

• Main Authors: María D. Bustamante-Rodríguez, Alberto A. Piedrahita-Ospina, Iliana M. Ramírez-Velásquez
• Format: Article
• Language: English
• Published: Instituto Tecnológico Metropolitano 2018-05-01
• Series: TecnoLógicas
• Subjects: Computational linguistics; text analysis; natural language processing; artificial intelligence; syntax
• Online Access: http://revistas.itm.edu.co/index.php/tecnologicas/article/view/788/916

Evaluating written texts is a task that mainly considers two aspects: syntactics and semantics. The first one focuses on the form of the text, and the second one, on its meaning. Conducting this task manually implies an effort in time and resources that can be reduced if part of the process is carried out automatically. According to the reviewed literature, there are different techniques for automatically correcting texts. One of them is the linguistic approach, which focuses on syntactic, semantic, and pragmatic elements. Likewise, this ongoing research is concerned with the automatic evaluation of syntactic errors in texts written in Spanish as a starting point to ensure coherence and cohesion in text composition, which may be useful in the academic environment. In order to carry out this study, a set of texts by students enrolled in an academic program was collected and analyzed by applying natural language processing and machine learning techniques. Additionally, the content of the corpus was manually corrected to compare the results of both methods, and correspondence was established between them. For this reason, it was concluded that the automatic method supports the syntactic correction process of a text written in Spanish.