Sentiment analysis for Indonesia hotel services review using optimized neural network

Abstract

In this paper, the sentiment model for the classification of review of hotel services in Indonesia is proposed to get a recommendation supporting the decision to search for the best hotel as the background in this study. The neural network model is applied to get the best model in this sentiment analysis. The stages of the research were carried out by first preprocessing the text data obtained through feature selection, in addition to optimizing the selection of attributes and optimizing the parameters carried out to improve the accuracy of the obtained classification. The genetic algorithm is proposed for parameter optimization due to limitations in neural networks in determining parameter values. Based on the results of the experiment, in the neural network model after optimization, the best accuracy rate is 88.99%.