Study that explores the contribution of opinion mining in customer service 2.0 databases extracted from Facebook, in order to measure consumers (in)satisfaction. The research aims to evaluate and propose tools used in the process of discovering knowledge in text and employing them in opinion mining at sentence level, as well as to use an opinion analysis methodology, choosing the NetVizz tool to extract the database from Facebook, the Microsoft Excel® for selection and reduction of data, Python codes for cleaning and transformation and the Semantria tool for text analysis. The pre-processed base for opinion mining is submitted to Naive Bayes, SMO and J48 algorithms in the Weka tool. The Research shows satisfactory results in opinion mining, with the best hit rate obtained in the SMO algorithm. Future works are proposed on customer service bases, for further comparison of obtained results on customer service 2.0 databases, seeking improvements.
Sancliment Iglesias, Bel. Luis and Tsunoda, Dr.ª Denise Fukumi, "Opinion Mining in Social Media as a Tool to Measure Consumer (In)Satisfaction" (2020). ISLA 2020 Proceedings. 18.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.