<![CDATA[Sentiment analysis to improve the quality of M-Passport Immigration services. The type of research to be conducted is descriptive research. Descriptive research is research that describes or explains a particular phenomenon, condition, or object systematically, factually, and accurately. This study collects information to answer the researcher's questions by paying attention to aspects obtained from many research data, so that it can describe a condition, event, or phenomenon specifically and sequentially. Therefore, this study wants to describe specifically and obtain facts about the sentiment of M-Papsor application users through reviews on the Google Play Store platform. The data source in this study is secondary data. Secondary data is taken from user reviews of the M-Passport application on the Google Play Store starting from June 1, 2022 to June 30, 2024 using the scrapping method. Based on the labeling process with the help of Python on the Google Collaboritory platform, the distribution of review data on the M-Passport application was obtained with positive user opinions of 12.32% and negative user opinions of 87.67%. Based on sentiment analysis using the Support Vector Machine (SVM) and Naïve Bayes Classifier (NBC) algorithms, the best accuracy results were obtained through the SVM method with an accuracy of 98%, precision 97%, recall 98%, and f-measure 97% while the NBC method obtained an accuracy of 87%, precision 88%, recall 84%, and f-measure 86%. Where, from these results it can be interpreted that the Support Vector Machine (SVM) algorithm has good performance in classifying sentiment on the M-Passport application. Based on the identification of the service quality variable in the negative sentiment class, the reason users gave the most negative comments was because of the reliability factor which obtained the highest percentage of 43.5%, the responsiveness factor of 27.6% empathy of 11.2%, tangibility of 8.9% and assurance of 8.7%.]]>
Copyrights © 2024
