Sentiment Analysis of JNT Express, JNE and Sicepat Customer Satisfaction on Twitter Using Multinomial Naïve Bayes Method
Zahra Fikri Ayu Nirwana, M Yoka Fathoni · 2024
The number of online buying and selling transactions through e-commerce as well as the need to deliver personal goods has led to an increase in the number of people using goods delivery services. Goods delivery service companies that are a favorite among Indonesians are JNT Express, JNE and Sicepat. Currently, these companies have been actively using various communication media, including Twitter to interact with their customers. Customer opinions about delivery services often appear on Twitter and sometimes become the main issue that steals the attention until it becomes a headline and becomes a hot conversation among the public. Analyzing customer sentiment through comments on Twitter is very important for JNT Express, JNE, and Sicepat. The Multinomial Naïve Bayes method was used to explore customers' views on their delivery services. A total of 1069 comment data was obtained using Google Collaboratory with the help of tweet-harvest from January 1, 2023 to December 31, 2023. The sentiment classification process was performed by utilizing Naïve Bayes classification to identify whether the reviews are positive, negative, or neutral. The classification results of the dataset with a composition of 75% of the training data and 25% of the test data on the JNT Express expedition obtained an accuracy result of 80%, JNE obtained an accuracy result of 75% and Sicepat 67%. For datasets with 80% and 20% composition, JNT Express obtained an accuracy of 77%, JNE 74% and Sicepat 65%.