Abstract. In this paper, we compare lexicon-based and machine learning-based approaches to define the subjectivity of tweets in Portuguese. We tested SentiLex and WordAffectBR lexicons, and Sequential Machine Optimization and Naive Bayes algorithms for this task. In our study, we used the Computer-BR corpus that contains messages about the technology area. We obtained better results using the Comprehensive Measurement Feature Selection method and the Sequential Machine Optimization algorithm as the classifier. We achieved considerable accuracy when we included the polarities of words in the vector space model of tweets.
This paper describes an ongoing research. We are building a chatterbot for education, in Portuguese, which uses conceptual lattices to generate answers for the students. We automatically build conceptual lattices from instructional materials about Artificial Intelligence. In this paper we describe how the lattices are generated for chatbot system.Resumo. Este artigo descreve uma pesquisa em andamento. Nós estamos construíndo um chatterbot para a educação, em Português, que usa reticulados conceituais para gerar as respostas para os estudantes. Nós geramos automaticamente esses reticulados conceituais a partir de materiais instrucionais sobre Inteligência Artificial. Esse artigo nós descrevemos como os reticulados do chatterbot são gerados.
This paper presents an investigation about concepts extraction from texts using clustering algorithms. We applied a hybrid approach to select feature candidates and the CLUTO tool to support the process of clustering of terms. The analysis of identified concepts was manual. The details and preliminaries results of this approach for portuguese texts are discussed.
RESUMOEste artigo apresenta um estudo sobre extração de conceitos a partir de textos usando algoritmos de agrupamento. Utilizamos uma abordagem hibrida para selecionar os termos candidatos e a ferramenta CLUTO para apoiar o processo de clusterização de termos. A análise dos conceitos identificados foi feita manualmente. O artigo apresenta o detalhamento desse processo e discute resultados preliminares em textos da língua portuguesa.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.