Pengaruh Faktor Eksternal Terhadap Prediksi Kunjungan Wisatawan Mancanegara Ke Indonesia Berbasis Model Transformer

Kusnadi, Kayla Kirani (2026) Pengaruh Faktor Eksternal Terhadap Prediksi Kunjungan Wisatawan Mancanegara Ke Indonesia Berbasis Model Transformer. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 5026221111-Undergraduate_Thesis.pdf] Text
5026221111-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only

Download (9MB) | Request a copy

Abstract

Pemerintah Indonesia telah menetapkan target ambisius untuk mencapai tingkat pertumbuhan ekonomi nasional sebesar 8% per tahun pada tahun 2029, dengan sektor pariwisata diidentifikasi sebagai katalisator utama. Dengan target pendapatan devisa yang meningkat menjadi USD 22,1 miliar pada tahun 2025, dan terus meningkat menjadi USD 24,7 miliar pada tahun 2026, diperlukan sistem prediksi pengunjung yang adaptif yang mengintegrasikan faktor eksternal seperti cuaca, kurs mata uang, dan perilaku pencarian daring. Studi ini membahas bagaimana faktor eksternal mempengaruhi jumlah kunjungan wisatawan mancanegara di Bali, DI Yogyakarta, dan DKI Jakarta. Solusi yang diusulkan untuk masalah ini adalah pengembangan model prediksi kunjungan wisatawan mancanegara berbasis arsitektur Transformer yang mengintegrasikan variabel eksternal multivariate. Evaluasi uji Diebold-Mariano menunjukkan pergeseran pola musiman menjadi lebih fleksibel, ditandai dengan kegagalan model kalender statis. Dalam hal kausalitas, variabel makroekonomi terbukti tidak signifikan (noise). Sebaliknya, minat digital muncul sebagai prediktor jangka panjang yang lebih baik, namun efektivitasnya bergantung pada seleksi fitur yang ketat. Analisis menunjukkan bahwa variabel iklim memiliki relevansi taktis dan strategis di DKI Jakarta dan Bali. Temuan tugas akhir ini mencakup strategi adaptif per wilayah, dimana pendekatan hibrida untuk Bali, koreksi sinyal multivariate untuk mengatasi gangguan struktural di DI Yogyakarta, dan dominasi stabilitas model univariate di DKI Jakarta. Temuan ini menunjukkan bahwa akurasi prediksi bergantung pada kesederhanaan dan kurasi data yang teliti.
=======================================================================================================================================
The Indonesian government has set an ambitious target of achieving an annual national economic growth rate of 8% by 2029, with the tourism sector identified as the main catalyst for this ambitious goal. With foreign exchange earnings targets increasing to USD 22.1 billion in 2025 and continuing to rise to USD 24.7 billion in 2026, an adaptive visitor prediction system is needed that integrates external factors such as weather, currency exchange rates, and online search behavior. This study discusses how external factors affect the number of foreign tourist visits to Bali, DI Yogyakarta, and DKI Jakarta. The proposed solution to this problem is the development of a foreign tourist visit prediction model based on a Transformer architecture that integrates multivariate external variables. Evaluation of 380 scenarios using the Diebold-Mariano test shows a shift in seasonal patterns to become more flexible, marked by the failure of static calendar models. In terms of causality, macroeconomic variables proved to be insignificant (noise). Conversely, digital interest emerged as a better long-term predictor, but its effectiveness depends on strict feature selection. The analysis shows that climate variables have tactical and strategic relevance in DKI Jakarta and Bali. The findings of this final project include adaptive strategies by region, where a hybrid approach is used for Bali, multivariate signal correction is used to overcome structural disturbances in DI Yogyakarta, and univariate model stability dominates in DKI Jakarta. These findings show that prediction accuracy depends on simplicity and careful data curation.

Item Type: Thesis (Other)
Uncontrolled Keywords: Wisatawan Mancanegara, Prediksi Pariwisata, Transformer Encoder, Prediksi Deret Waktu, Inbound Tourism, Tourism Prediction, Transformer Encoder, Time Series Prediction
Subjects: Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry)
Q Science > QA Mathematics > QA278.5 Principal components analysis. Factor analysis. Correspondence analysis (Statistics)
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Q Science > QA Mathematics > QA278 Cluster Analysis. Multivariate analysis. Correspondence analysis (Statistics)
T Technology > T Technology (General) > T57.5 Data Processing
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis
Depositing User: Kusnadi Kayla Kirani
Date Deposited: 20 Jan 2026 03:06
Last Modified: 20 Jan 2026 03:06
URI: http://repository.its.ac.id/id/eprint/129787

Actions (login required)

View Item View Item