Sentiment Analysis of Mobile App Reviews on Google App Stores Alifia Puspaningrum, Munengsih Sari Bunga, Iryanto
Politeknik Negeri Indramayu
Abstract
Software maintenance is a prior process in Software Development Live Cycle. Google App Store has been already supported software developer to do their maintenance process by collecting user reviews. By analyzing these reviews, software developers can analyze user sentiment towards their applications. Sentiment analysis is one method for identifying negative of positive opinions. This paper classify user satisfactions sentiment in Mobile App Reviews by comparing some classification method such as Support Vector Machine (SVM), Naive Bayes (NB), and Random Forest (RF). Features used for the classification is TF-IDF. Performance of the classification is evaluated by using Precision, Recall, F-Measure. Experimental result show that Naive Bayes achieve the highest performance.
Keywords: Sentiment Analysis, Mobile App Review, Machine Learning
