Title :
Comparison of different algorithms for sentiment classification
Author :
Ciric, Miroslav ; Stanimirovic, Aleksandar ; Petrovic, Nikola ; Stoimenov, Leonid
Author_Institution :
Fac. of Electron. Eng., Univ. of Nis, Niš, Serbia
Abstract :
Sentiment classification has various applications and information from social networks can be especially useful. In this paper we perform sentiment classification of Twitter messages, so called tweets. We compare several machine learning classification algorithms and try to improve results by using processing pipes that extract meaningful features and remove noise.
Keywords :
classification; learning (artificial intelligence); social networking (online); Twitter messages; machine learning classification algorithms; processing pipes; sentiment classification; social networks; tweets; Accuracy; Classification algorithms; Entropy; Machine learning algorithms; Nickel; Training; Twitter; Machine learning; Sentiment classification; Twitter;
Conference_Titel :
Telecommunication in Modern Satellite, Cable and Broadcasting Services (TELSIKS), 2013 11th International Conference on
Conference_Location :
Nis
Print_ISBN :
978-1-4799-0899-8
DOI :
10.1109/TELSKS.2013.6704442