Um Avaliador Automático de Redações

• Ano de defesa: 2021
• Autor(a) principal: Silva Junior, James Alves da
• 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: Avaliação automática de redações; processamento de linguagem natural; ENEM; subject.br-rjbn; Ciência da Computação
• Link de acesso: http://repositorio.ufes.br/handle/10/15483

In Brazil, admission to higher education is necessary to pass the Exame Nacional do Ensino Médio (ENEM) conducted by the Instituto Nacional de Estudos e Pesquisas Educacionais Anísio Teixeira (INEP). A final grade of the exam directly interferes with student life, since the final grade of the exam makes it possible to enter higher education institutions. A note also interferes with student funding. The student’s point of view a low grade is the loss of opportunities. Writing is an important part of the composition of the exam score, since the writing score for students with zero will be automatically eliminated by the Sistema de Seleção Unificada (SISU) for admission to public institutions of higher education and also with the student credit process. . In view of the importance of writing in the composition of the final grade, explore two Competencies that are evaluated in the writing of the ENEM/INEP model. This work seeks to create an Automatic Essay Evaluation model (AES) that aims to imitate human evaluators and give grades in essays. This area has been around for 50 years and the English language has made many consistent advances in estimates with TOEL, Cambrige. The automatic writing correction system is not widely adopted in Portuguese without support for learning and even as an evaluation system. Research on the topic has grown since there is an intention to make the assessment of ENEM completely computerized. Other discussions on the subject of writing are about a dichotomy between the evaluators when evaluating the essays and also how the model can provide feedbacks to improve student performance when writing as essays.