<![CDATA[Reviews of a service, especially hotel services, have an important role to play in consumer decisions. Tripadvisor is a guide and reference platform for travelers in finding information about the hotel services in various countries. There are many reviews about hotels on the platform so that readers are difficult to make decisions so it is necessary to conduct a sentiment analysis that aims to dig up information from existing reviews. The initial stage is by labeling (positive, neutral, negative) to the review. Then the preprocessing stage so that the review can be easily processed, then from that stage continued weighting using Term Frequency - Inverse Document Frequency (TF-IDF) using the best parameters, after weighting the data, then the next is the distribution of data into training data, validation and test. The data are entered into the machine learning process using Support Vector Machine (SVM) and obtained the accuracy of the model by 85%. For testing scenarios if not using slang handling get F1-Score by 80% and if not using stopword get F1-Score by 82%. On the evaluation of the performance of the model using K-Fold obtained the best results on the Fold-7 with a precision value of 87%, recall 86%, F1-score 86%, and accuracy of 87%.]]>
Copyrights © 2022
