Sentiment classification of micro‐blog comments based on Randomforest algorithm

• Published in: Concurrency and computation Vol. 31; no. 10
• Main Authors: Liu, Xiao‐Qin, Wu, Qiu‐Lin, Pan, Wen‐Tsao
• Format: Journal Article
• Language: English
• Published: Hoboken Wiley Subscription Services, Inc 25.05.2019
• Subjects: Accuracy; Algorithms; Attitudes; Bayesian analysis; Cell phones; Cellular telephones; Classification; Classifiers; Consumers; Data mining; Machine learning; Microblogs; micro‐blog comments; mobile phone brands; Noise reduction; R language software; Randomforest; Sentiment analysis; sentiment classification; Support vector machines; Thesauri
• ISSN: 1532-0626; 1532-0634
• DOI: 10.1002/cpe.4746

Summary Sentiment classification of Micro‐blog comments aims to distinguish consumers' attitudes toward a certain brand or an event. A great number of research studies related to sentiment analysis have been conducted based on machine learning methods, such as Support Vector Machine and Naive Bayes, to build the classifier, and some other research studies tried to build one based on emotional thesaurus. However, the rate of accuracy is yet to improve. In order to improve the accuracy rate further, this thesis bases on the Randomforest algorithm to construct an emotional tendency classifier to help the mobile phone service providers to further understand consumers' attitudes toward their brand. The Randomforest algorithm is a bagging algorithm of the ensemble learning with some weak classifiers, which works very well in anti‐noise and reducing the overfitting problem. Finally, a classifier with an 83% accuracy rate for sentiment classification of mobile phone brands on Micro‐blog comments was built based on Randomforest by using R language software.