Detección de estados de ánimo mediante sentiment analysis en hispanohablantes

• Autor: Morzán Fuentes, Stephany Alessandra
• Formato: tesis de grado
• Fecha de Publicación: 2019
• Institución: Universidad de Lima
• Repositorio: ULIMA-Institucional
• Lenguaje: español
• OAI Identifier: oai:repositorio.ulima.edu.pe:20.500.12724/11318
• Enlace del recurso: https://hdl.handle.net/20.500.12724/11318; http://doi.org/10.26439/ulima.tesis/11318
• Nivel de acceso: acceso abierto
• Materia: Lenguaje y emociones; Análisis de texto (Minería de datos); Proceso en lenguaje natural (informática); Lingüística computacional; Hispanoamericanos; Language and emotions; Text Analysis (Data Mining); Natural language processing (computing); Computational linguistics; Hispanic americans; https://purl.org/pe-repo/ocde/ford#2.02.04

The following research shows the elaboration and validation of a computational model designed to detect Spanish speakers’ moods from answers entered by these users in text format. It was used the Keyword Spotting Technique (KST) for text treatment, and Sentiment Analysis based on only Natural Language Processing (NLP) concepts for text analysis. A mobile application with chatbot interface and a bot that invites the user to give details about their mood through questions validated by an expert, are the tools used to collect all the conversations. As a result of the pilot test of 30 conversations, and after the final test in the sample of 49 conversations, there is a correct classification of 70% of the cases, as well as a proposal for the treatment of texts in Spanish.