Opinion analysis for twitter and arabic tweets: a systematic literature review
Mnahel Ahmed Ibrahim, Naomie Binti Salim · Journal of Theoretical and Applied Information Technology · 2013
The objective of this paper is to present the current evidence relative to twitter opinion mining in general and also the current state of Arabic tweets’ opinion mining. The researcher performed a systematic literature review (SLR) to investigate features and methods used for twitter opinion mining and if those features and methods have been used for Arabic tweets opinion mining. Sixty five papers were used in our synthesis of evidences. Results showed that n-grams features are the most features used for twitter sentiments analysis and also for Arabic tweets. The most common methods used for twitter sentiments analysis is the Lexical based classification using Naive Bayes (NB) and Support Vector Machines (SVM), which are also used for Arabic tweets. In addition, evidence related to subjectivity and opinion target for twitter are highlighted. The results of this SLR show gaps in the research field: first, the lack of studies focusing on multilingual twitter sentiments analysis. Second, the lack of studies that investigate Arabic tweet opinion target. The third is the lack of studies investigating Arabic tweet subjectivity.