Use este identificador para citar ou linkar para este item: http://repositorio.ufla.br/jspui/handle/1/32175
Título: BigFeel: um ambiente de processamento distribuído para integração de métodos de análise de sentimentos
Autores: Pereira, Denilson Alves
Ribeiro, Leonardo Andrade
Zambalde, André Luiz
Palavras-chave: Análise de sentimento
Aprendizagem de máquina
Processamento de linguagem natural
Sentiment analysis
Machine learning
Natural language processing
Hadoop
Spark
Big data
Data do documento: 17-Dez-2018
Editor: Universidade Federal de Lavras
Citação: FERREIRA, R. S. BigFeel: um ambiente de processamento distribuído para integração de métodos de análise de sentimentos. 2018. 101 p. Dissertação (Mestrado em Ciência da Computação)–Universidade Federal de Lavras, Lavras, 2018.
Resumo: Sentiment analysis has been the main focus of plenty of research efforts, particularly justified by its commercial significance, both for consumers and businesses. Thus, many methods have been proposed, and the main ones have been compared in terms of effectiveness. Nonetheless, the literature is deficient when it comes to assessing the efficiency of these methods for processing large volumes of data, which are generated at great speed, volume and variety, known as Big Data. The present work presents an approach for integrating methods of sentiment analysis in order to process large volumes of data in a distributed environment, using both the Apache Hadoop and Spark platforms. A distributed application prototype was developed, named BigFeel, which supports the use of 22 methods of sentiment analysis, as well as some methods of natural language processing and textual preprocessing in large volumes of data. BigFeel offers services tailored to the use of computer networks, local and web, as well as offering an API for Scala/Java developers. The efficiency of the integrated methods was evaluated experimentally, demonstrating gain in comparison to the execution in the non-distributed implementation of the methods. Using the features offered by BigFeel, a case study of detection of innovation suggestions based on product and service reviews is also presented.
URI: http://repositorio.ufla.br/jspui/handle/1/32175
Aparece nas coleções:Ciência da Computação - Mestrado (Dissertações)



Os itens no repositório estão protegidos por copyright, com todos os direitos reservados, salvo quando é indicado o contrário.