This paper focuses on how naive Bayes classifiers work in opinion mining applications. The first question asked is what are the feature sets to choose when training such a classifier in order to obtain the best results in the classification of objects (in this case, texts). The second question is whether combining the results of Naive Bayes classifiers trained on different feature sets has a positive effect on the final results. Two data bases consisting of negative and positive movie reviews were used when training and testing the classifiers for testing purposes. © 2012 Springer-Verlag.
CITATION STYLE
Dinu, L. P., & Iuga, I. (2012). The naive Bayes classifier in opinion mining: In search of the best feature set. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7181 LNCS, pp. 556–567). https://doi.org/10.1007/978-3-642-28604-9_45
Mendeley helps you to discover research relevant for your work.