A study of the effects of preprocessing strategies on sentiment analysis for Arabic text

Rehab Duwairi, Mahmoud El-Orfali
2014 Journal of information science  
Sentiment analysis has drawn a considerable interest among researchers due to the realization of its fascinating commercial and business benefits. This paper deals with sentiments analysis in Arabic text from three perspectives. Firstly, several alternatives of text representation were investigated. In particular, the effects of stemming, feature correlation, and n-gram models for Arabic text on sentiment analysis were investigated. Secondly, the behavior of three classifiers, namely, SVM,
more » ... Bayes, and K-nearest neighbor classifiers, with sentiment analysis was investigated. Thirdly, the effects of the characteristics of the dataset on sentiment analysis were analyzed. To this end, we have applied the techniques proposed in this paper to two datasets; one was prepared inhouse by the authors and the second one is freely available online. All the experimentation was done using Rapidminer 1 . The results show that our selection of preprocessing strategies on the reviews increases the performance of the classifiers.
doi:10.1177/0165551514534143 fatcat:zcj66yfyd5a5lfyoa3jy6g3bui