Implementation Of K-Nearest Neighbor Algorithm With SMOTE For Hotel Reviews Sentiment Analysis

Indonesia has considerable tourism development potential, this phenomenon is in accordance with the number of foreign tourist visits to from January September 2022 recorded by Badan Pusat Statistik many as 2,397,181 visitors. This research focuses on super-priority destinations Labuan Bajo, East Nusa Tenggara, based government's plan that focus developing destination increase hotel meet need for an additional 2,000 rooms. Thus, available rooms are still limited. Then choose a November 2021 survey Populix website, 76% 1,012 respondents chose book hotels online majority using Traveloka website. However, making decisions choosing reviews feature website raises various problems, such biased information and even rating values ​​given do not match submitted. So users know what becomes perception positive negative ratings, it necessary in-depth satisfaction factors find out sentiments study uses k-nearest neighbor algorithm SMOTE objects three most popular Bajo. Data testing value k = 3 so produces accuracy 87.71% - 93.47% maximum error tolerance 12.29%. In addition, performance results validated appropriate AUC value, namely good classification category.