Comparative Sentiment Analysis of Delivery Service PT.POS Indonesia and J&T Express on Twitter Social Media Using The Support Verctor Machine Algorithm

Abstract

Based on a survey conducted by the Top Brand Award in the courier service category, the J&T Express company is in the highest position from 2018 to 2021 beating Pos Indonesia. Social media Twitter is a place often used by customers to submit complaints and opinions regarding the services of a company. The method used to determine the tendency of the views to contain positive or negative sentiments is sentiment analysis. Sentiment analysis will classify the polarity of the text in sentences or documents to determine whether the opinions expressed are positive or negative. This study uses the Support Vector Machine (SVM) algorithm. The results of the user tweet data used are as many as 1000 data with details of data 206 (20.6%) have positive sentiments and 794 (79.4%) have negative sentiments. In the Pos Indonesia tweet data, 110 positive sentiment data were obtained, while the positive sentiment data in the J&T Express tweet data was 96 data. This shows that the Pos Indonesia delivery service has better customer service than J&T Express. The highest level of accuracy using the SVM algorithm in classifying sentiment is 80.14% with a comparison of 70% training data and 30% test data with an average precision of 90%, an average recall of 51.74%, and an average f-measure of 47.80%.