The development of the Mandalika National Strategic Tourism Area (KSPN) faces challenges in maintaining a balance between the various components of tourism infrastructure. This study aims to analyse tourist sentiment from the Google Maps and TripAdvisor platforms and to evaluate infrastructure based on the 4A components, namely ‘Attraction’, ‘Accessibility’, ‘Amenity’ and ‘Ancillary’. A combined method using an “Explanatory Sequential” design was employed, involving text processing, TF-IDF weighting, SMOTE data balancing, and classification using the Support Vector Machine (SVM) and Naïve Bayes algorithms. A total of 1,600 reviews were analysed and examined in greater depth through observation and interviews. The test results showed that the SVM algorithm had an accuracy of 77.43 per cent, higher than that of the Naïve Bayes algorithm at 72.10 per cent. The sentiment distribution was dominated by positive sentiment at 50.8 per cent, negative at 40.7 per cent, and neutral at 8.5 per cent. The 4A evaluation revealed that the area’s strengths lie in ‘Attraction’, whilst complaints were concentrated on ‘Accessibility’ - such as poorly lit roads - ‘Amenity’ - such as the toilet crisis at Bukit Merese, and ‘Ancillary’ issues, such as the coercion of transactions by local traders. This study concludes that the equitable distribution of basic facilities in buffer zones is essential for inclusive and sustainable tourism development.
Copyrights © 2026