User Satisfaction Levels Sentiment Analysis Toward Goods Delivery Service On Twitter Using Support Vector Machine Algorithm (SVM)

摘要

The data amount is experiencing rapid growth in this era. Data can be in the form of text, images, sound or video. Social media has become one of the factors of data growth. Every person can express their feelings on social media from these opinions can be analyzed. In this study sentiment analysis uses the support vector machine algorithm. The first step is crawling data using the Twitter API with keywords. After collecting data, the preprocessing process is carried out, after the preprocessing process the feature is retrieved for each tweet, the features obtained are then collected into a feature list. The feature list is then transformed into a feature vector in binary form and then transformed using the TF-IDF method. The dataset consists of 2 data, namely training and testing. The training is labelled manually. To test the performance of the algorithm used the K-Fold Cross Validation method. The test results are obtained an average accuracy of 98% with the composition of training data and testing data. From these results, the Support Vector Machine method can be used for sentiment classification of JNE twitter data.