Mineração de Texto para a Análise do Perfil Emocional de Usuários de Jogo Empático

• DOI: 10.14210/cotb.v12.p370-377
• Publication Date: 2021-05-04T14:34:08Z
• Authors: Leonardo Dias Mar...; Fabíola Pantoja O...
• Keywords: Kernel (algebra)

Daily, a large amount of data circulates on the Internet, producing lot information in form images, videos and texts. Then, it is necessary to analyze extract these automatically. Therefore, this work presents case study that applies text mining emotional sentimental profiles from comments Last Day June game users, where results extracted analysis sentiments were presented. Three classification algorithms used: Naive Bayes, Support Vector Machine (SVM) K-Nearest Neighbors (KNN) predict class elements according emotions or feelings identified analysis. As result, SVM with radial kernel was one best accuracy, 79%, followed by KNN 3 closest neighbors, 75%, finally, 62%.