Proposta de um sistema automático de avaliação de redações do ENEM, foco na competência 1: demonstrar domínio da modalidade escrita formal da língua portuguesa

• Ano de defesa: 2017
• Autor(a) principal: Almeida Júnior, Celso Romão Cardoso de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Espírito Santo; BR; Mestrado em Informática; Centro Tecnológico; UFES; Programa de Pós-Graduação em Informática
• Palavras-chave em Português: Natural language processing; Machine learning; Artificial intelligence; Essay; Processamento de linguagem natural; Aprendizagem de máquina; ENEM; Redação; Aprendizado do computador; Inteligência artificial; Processamento de linguagem natural (Computação); Ciência da Computação; 004
• Link de acesso: http://repositorio.ufes.br/handle/10/10985

Automatic essay assessments are widely practiced in English, but in Portuguese, we cannot say the same. Writing is one of the competencies required by the National High School Examination (ENEM), gateway to most universities in Brazil. The high cost and the large number of professionals working in the correction process of the ENEM essays are some of the factors that motivate research in the area of automatic essay evaluation. This work presents a strategy to improve the evaluator’s productivity, reducing the effort in 20% of the time, evaluating Competence 1, one of the five competences evaluated in the essays of ENEM. For this, we propose the construction of a system of automatic evaluation of essays, in Competence 1 of the ENEM; demonstrate mastery of the formal written form of the Portuguese language. In the construction of the system, we use Natural Language Processing techniques and tools, in the preprocessing stage of the essays as well as Machine Learning techniques in the stages of selection of characteristics and prediction of the grades. The results of the experiments carried out with the UOL website show that the system is able to support the ENEM essay evaluator with an absolute mean error of 0.2354 in 2.0 compared to the scores attributed by the site experts.