Implementasi Particle Swarm Optimization (PSO) pada Analysis Sentiment Review Aplikasi Trafi menggunakan Algoritma Naive Bayes (NB)

Abstract

Abstract - The development of transportation applications is now getting bigger so that many vendors compete for business in creating transportation mode applications, starting from the quality and quantity so that it is often questioned. With this, the researcher held a transportation application called Trafi to get opinions or comments on applications from people who had used the application and poured it into online media. Of the many comments reviewed to obtain a set of positive and negative forms of data from the text that the researcher will process. For classification data using Naïve Bayes (NB), NB is one of the most popular algorithms for pattern recognition. Apart from simplicity, the Naive Bayes classifier is a popular machine learning technique for text classification, Particle Swarm Optimization (PSO) which combines with the Naive Bayes classification to improve performance. Before use, optimization with PSO in the data set accuracy obtained was 69.50% and after the combination of Naive Bayes and PSO accuracy was 72.34%. Use PSO and Naïve Bayes according to the concept of text mining which aims to find patterns that exist in text, the activity carried out by text mining here is text classification.Abstract - The development of transportation applications is now getting bigger so that many vendors compete for business in creating transportation mode applications, starting from the quality and quantity so that it is often questioned. With this, the researcher held a transportation application called Trafi to get opinions or comments on applications from people who had used the application and poured it into online media. Of the many comments reviewed to obtain a set of positive and negative forms of data from the text that the researcher will process. For classification data using Naïve Bayes (NB), NB is one of the most popular algorithms for pattern recognition. Apart from simplicity, the Naive Bayes classifier is a popular machine learning technique for text classification, Particle Swarm Optimization (PSO) which combines with the Naive Bayes classification to improve performance. Before use, optimization with PSO in the data set accuracy obtained was 69.50% and after the combination of Naive Bayes and PSO accuracy was 72.34%. Use PSO and Naïve Bayes according to the concept of text mining which aims to find patterns that exist in text, the activity carried out by text mining here is text classification.Keywords: Sentiment Analysis, Android Appstore Product Review, Naive Bayes Algorithm