Hidayaturrohman, Bilal (2025) Peramalan Jumlah Kedatangan Turis Menggunakan SARIMA-CNN-BiLSTM dengan Perbandingan Data Media Sosial dan Data Penumpang Transportasi. Masters thesis, Institut Teknologi Sepuluh Nopember.
![]() |
Text
6026231010-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (3MB) | Request a copy |
Abstract
Industri pariwisata merupakan sektor ekonomi yang berperan penting dalam perekonomian suatu negara, termasuk Indonesia. Penelitian ini membandingkan efektivitas data penumpang transportasi dan data sosial media dalam memprediksi kedatangan wisatawan serta mengusulkan model gabungan SARIMA-CNN-BiLSTM untuk meningkatkan akurasi prediksi. Hasil penelitian menunjukkan bahwa data penumpang transportasi lebih akurat dibandingkan data sosial media, dengan nilai R-Square yang lebih tinggi, tetapi kombinasi keduanya meningkatkan akurasi prediksi, terutama di Jakarta. Metode SARIMA-CNN-BiLSTM terbukti lebih unggul dibandingkan metode lain seperti SARIMA, CNN-BiLSTM, maupun SARIMA-CNN-LSTM, karena mampu menangkap pola musiman dan mengenali pola kompleks dalam data, meskipun pada data multivariate, CNN-BiLSTM menunjukkan performa lebih baik. Prediksi kedatangan wisatawan untuk tahun 2024 menunjukkan pola pertumbuhan moderat pada data penumpang transportasi, sementara data sosial media mencerminkan potensi fluktuasi yang lebih dinamis. Oleh karena itu, data sosial media disarankan sebagai salah satu sumber utama dalam peramalan wisatawan, serta pengembangan metode SARIMA-CNN-BiLSTM untuk meningkatkan performa model multivariate. Penelitian lebih lanjut dapat mengeksplorasi lebih banyak platform media sosial dan menambahkan variabel eksternal.
=====================================================================================================================================
The tourism industry is a crucial economic sector that plays a significant role in a country's economy, including Indonesia. This study compares the effectiveness of transportation passenger data and social media data in predicting tourist arrivals and proposes a hybrid SARIMA-CNN-BiLSTM model to enhance prediction accuracy. The findings indicate that transportation passenger data is more accurate than social media data, with a higher R-Square value, but combining both data sources improves prediction accuracy, especially in Jakarta. The SARIMA-CNN-BiLSTM method outperforms other approaches, such as SARIMA, CNN-BiLSTM, and SARIMA-CNN-LSTM, due to its ability to capture seasonal patterns and recognize complex data structures, although for multivariate data, CNN-BiLSTM performs better. The 2024 tourist arrival forecast shows a moderate growth pattern in transportation passenger data, while social media data reflects more dynamic fluctuations. Therefore, social media data is recommended as a key source for tourist arrival forecasting, along with further development of the SARIMA-CNN-BiLSTM method to improve multivariate model performance. Future research can explore more social media platforms and incorporate external variables to enhance predictive accuracy.
Item Type: | Thesis (Masters) |
---|---|
Uncontrolled Keywords: | SARIMA-CNN-BiLSTM, Peramalan, Kedatangan Turis, Sosial Media, SARIMA-CNN-BiLSTM, Forecasting, Tourist Arrival, Social Media |
Subjects: | T Technology > T Technology (General) > T174 Technological forecasting T Technology > T Technology (General) > T57.5 Data Processing |
Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 59101-(S2) Master Thesis |
Depositing User: | Bilal Hidayaturrohman |
Date Deposited: | 01 Feb 2025 15:20 |
Last Modified: | 01 Feb 2025 15:20 |
URI: | http://repository.its.ac.id/id/eprint/117607 |
Actions (login required)
![]() |
View Item |