Xavier Júnior, João CarlosPorto, Diego Rolim2022-07-052022-07-052022-06-15PORTO, Diego Rolim. Classificação automática de documentos baseada em mineração de texto e processamento de linguagem natural. 2022. 14 f. Trabalho de Conclusão de Curso (Residência em Tecnologia da Informação) - Instituto Metrópole Digital, Universidade Federal do Rio Grande do Norte, Natal, 2022.https://repositorio.ufrn.br/handle/123456789/48335The manual classification of documents represents, in most cases, a slow and demanding process since it is based on reading part of the documents. Based on this fact, the main objective of this work is to carry out a study of different Text Mining and Natural Language Processing (NLP) techniques for the automatic classification of documents related to the accountability of the city councils of the Rio Grande do Norte State. In this sense, we have chosen two methods found in the literature, as being: TF-IDF and Doc2Vec, because they have distinctive characteristics. In this context, it is important to analyze the performance of both methods, as well as the complexity in the construction of dictionaries to be used in the first, and the necessary training stage for the second. For this end, two sets of documents were created, one for training or creating dictionaries, and another for testing both methods. In this sense, the experimental results showed that the methodology based on Doc2Vec is more indicated to be used by the State's Court of Auditors. This result is justified by the accuracy of 100% obtained in the performed tests and due to better scalability of the implementations used in the method.Attribution 3.0 Brazilhttp://creativecommons.org/licenses/by/3.0/br/Mineração de textoProcessamento de linguagem naturalTF-IDFDoc2VecmasterThesis