Integrasi Algoritma Genetika Dan Information Gaint Untuk Menganalisis Sentimen Review Hotel Menggunakan Algoritma Naive Bayes

Abstract

Input and advice is one important part of the application site, in order to assess and improve a quality and quality, Reading reviews helps consumers choose the best hotels, help companies and developers to monitor user satisfaction to improve the quality and quantity of features and services, read as a whole and in manual can spend quite a long time, if read at a glance, the information is not delivered perfectly. This study analyzes the user sentiment Agoda Hotels by automatically classifying reviews for a positive or negative opinion. To improve the accuracy of Naïve Bayes methods Feature Selection, Information Gain and genetic algorithms. This model was evaluated using 10 Fold Cross Validation. Measurements were made with the Confusion Matrix and the ROC curve, comparing accuracy before and after the addition of feature selection methods. The results showed an increase in accuracy, 60.50% to 83.00%.