Analisis Sentimen Transportasi Online Menggunakan Ekstraksi Fitur Model Word2vec Text Embedding Dan Algoritma Support Vector Machine (SVM)

Abstract

In the era of society 5.0, information technology is growing rapidly, one of which is in the field of transportation. The phenomenon of online transportation services is becoming increasingly popular among the public. With this phenomenon, many people have opinions about online transportation services, both positive and negative comments. The purpose of this study was to conduct a sentiment analysis of comments or users of online transportation service applications on gojek and grab on the Google Play Store. The stages of this research process are data collection, data labeling, data preprocessing, feature extraction and sentiment classification using the Support Vector Machine (SVM) algorithm. Data collection is done by web scraping. Dataset labeling is divided into two classes, namely positive sentiment and negative sentiment. Word2Vec Text Embedding is used as a feature extraction model to represent words in vector form. The architecture of the word2vec model used is the skip-gram model. The Support Vector Machine (SVM) algorithm is used for the data classification process to determine the level of accuracy of the data sentiment used. The results of tests carried out on the classification of sentiment analysis in online transportation applications show quite good performance results, namely for the Gojek application to get an accuracy rate of 87%, a precision of 93%, and a re-call of 84%. While the Grab application gets an accuracy rate of 82%, precision of 89%, and re-call of 83%.