Sentiment Analysis on Umrah Packages Review in Malaysia

• Published in: Lecture Notes in Networks and Systems
• Main Author: Dewi D.A.; Kurniawan T.B.; Zakaria M.Z.; Kasim S.; Mustapa N.Q.
• Format: Conference paper
• Language: English
• Published: Springer Science and Business Media Deutschland GmbH 2024
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85200944133&doi=10.1007%2f978-3-031-66965-1_21&partnerID=40&md5=49b9c61f3deebc8df78f97b86e0634a1

Umrah is a well-known pilgrimage site where individuals from around the world come together to receive good fortune from God in the here and now. In Malaysia, there are numerous sorts of Umrah packages promoted by many firms such as Andalusia Travel & Tours, Tiram Travel & Tours, Tabung Haji, and beyond. Because there are so many organizations offering Umrah trips, there are issues when careless people or organizations encourage fraud in Umrah packages to benefit from it. In 2023, the Police in Malaysia have confirmed that they have received 399 allegations regarding suspected Umrah package scams by a local company. The reports include 1,614 pilgrims around the country who are believed to have lost approximately RM14 million. With the world's technological advancements happening at a quick pace, individuals are more comfortable sharing their ideas and opinions on touchy themes on social media, including the experience with the fraud Umrah package. Performing a classification study from these resources develops the sentiment analysis against the available Umrah package. This classification approach can be used in searching applications to assist people in finding better Umrah packages. In this paper, Naïve Bayes, Random Forest, and Support Vector Machine are the classifiers that were employed in the study. The outcomes demonstrate that, out of all the classifiers, the Support Vector Machine has the highest accuracy. The Support Vector Machine (SVM) had a 90.92% accuracy rate. Ninety percent of the results were precise. It displays 90% overall for the f1-score and 91% overall for the total recall. Compared to Naïve Bayes, SVM has greater accuracy. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.