1.png)
PREDIKSI KUNJUNGAN WISATAWAN DOMESTIK ABNORMAL DENGAN ADAPTIVE LIGHTGBM DAN DETEKSI ANOMALI
Abstract
Sektor pariwisata Indonesia menghadapi fluktuasi signifikan dalam kunjungan wisatawan, yang dipengaruhi oleh berbagai faktor ekonomi, lingkungan, dan sosial. Penelitian ini mengembangkan model Adaptive LightGBM with Anomaly Detection and Multi-Task Learning (ALAD-MTL) untuk memprediksi kunjungan wisatawan domestik tahunan di tingkat provinsi dan mengidentifikasi pola kunjungan yang tidak normal. Model ini mengintegrasikan data statistik multi-domain (ekonomi, lingkungan, transportasi, fiskal) dan data media sosial (Twitter) yang kaya sentimen. Data mentah diolah dan dinormalisasi ke dalam format panjang, dengan data Twitter menjalani ekstraksi fitur sentimen menggunakan model RoBERTa berbahasa Indonesia dan agregasi. Semua data kemudian digabungkan ke dalam master_df dengan granularitas provinsi-tahunan. Model LightGBM dioptimalkan melalui penyetelan hyperparameter dan dievaluasi menggunakan metrik MAE, RMSE, R-squared, dan MAPE. Deteksi anomali diimplementasikan berdasarkan residual Z-Score dan Isolation Forest. Multi-Task Learning disimulasikan dengan melatih model LightGBM terpisah untuk tugas-tugas terkait (misalnya, Tingkat Hunian Hotel Bintang, Jumlah Tamu Hotel Asing) menggunakan set fitur yang sama. Hasil menunjukkan model mencapai R-squared tinggi (misalnya, >0.96 untuk tugas utama) dengan MAPE yang kompetitif, menunjukkan kemampuan prediksi pola yang kuat.
Full Text:
PDF (Bahasa Indonesia)References
R. S. Jinan, “Dampak Industri Pariwisata Bagi Perekonomian,” tirto.id, 2023. https://tirto.id/dampak-industri-pariwisata-bagi-perekonomian-gSW6 (accessed May 12, 2025).
K. Wisnubroto, “Meneropong Tren Pariwisata 2025,” indonesia.go.id, 2025. https://indonesia.go.id/kategori/editorial/8899/meneropong-tren-pariwisata-2025?lang=1#:~:text=Hasil survei ahli pada 2023 dan 2024 menunjukkan,salah satu andalan dalam menggenjot pertumbuhan ekonomi nasional. (accessed May 12, 2025).
Muryani and M. Esquivias, “Sektor Pariwisata sebagai Mesin Baru Pertumbuhan Ekonomi,” news.unair.ac.id, 2021. https://news.unair.ac.id/id/2021/05/18/sektor-pariwisata-sebagai-mesin-baru-pertumbuhan-ekonomi/ (accessed May 12, 2025).
D. Yang et al., “Population-Wide Depression Incidence Forecasting Comparing Autoregressive Integrated Moving Average and Vector Autoregressive Integrated Moving Average to Temporal Fusion Transformers: Longitudinal Observational Study,” J. Med. Internet Res., vol. 27, no. 1, pp. 1–13, 2025, doi: 10.2196/67156.
Y. Altork, “Comparative analysis of machine learning models for wind speed forecasting: Support vector machines, fine tree, and linear regression approaches,” Int. J. Thermofluids, vol. 27, no. April, p. 101217, 2025, doi: 10.1016/j.ijft.2025.101217.
K. Hasanpouri, F. Shokoohi, and Z. Farrokhi, “SP-RF-ARIMA : A sparse random forest and ARIMA hybrid model for electric load forecasting,” Glob. Energy Interconnect., 2025, doi: 10.1016/j.gloei.2025.04.003.
M. Amrani et al., “Short-Term Spatial-Temporal Energy Forecasting in a Danish Dis- tribution Grid Using a Hybrid Transformer-Graph Neural Network Model,” Comput. Condens. Matter, 2020, doi: 10.1016/j.spes.2025.05.002.
A. Mumuni and F. Mumuni, “Automated data processing and feature engineering for deep learning and big data applications: A survey,” J. Inf. Intell., vol. 3, no. 2, pp. 113–153, 2024, doi: 10.1016/j.jiixd.2024.01.002.
H. Eldin et al., “Journal of Radiation Research and Applied Sciences Advanced time complexity analysis for real-time COVID-19 prediction in Saudi Arabia using LightGBM and XGBoost,” J. Radiat. Res. Appl. Sci., vol. 18, no. 2, p. 101364, 2025, doi: 10.1016/j.jrras.2025.101364.
A. Syamsuddin, A. C. Adhi, A. Kusumawardhani, T. Prahasto, and A. Widodo, “Predictive maintenance based on anomaly detection in photovoltaic system using SCADA data and machine learning,” Results Eng., vol. 24, no. July, p. 103589, 2024, doi: 10.1016/j.rineng.2024.103589.
H. Talebi, A. Khatibi, and V. Khatibi, “Journal of Open Innovation : Technology , Market , and Complexity Developing a hybrid machine learning model for employee turnover prediction : Integrating LightGBM and genetic algorithms,” J. Open Innov. Technol. Mark. Complex., vol. 11, no. 2, p. 100557, 2025, doi: 10.1016/j.joitmc.2025.100557.
A. Hussain, A. Saadia, and F. M. Alserhani, “Ransomware detection and family classification using fine-tuned BERT and RoBERTa models,” Egypt. Informatics J., vol. 30, no. January, p. 100645, 2025, doi: 10.1016/j.eij.2025.100645.
M. A. El-Omairi, M. El Garouani, and A. El Garouani, “Enhanced lithological mapping via remote sensing: Employing SVM, random trees, ANN, with MNF and PCA transformations,” Egypt. J. Remote Sens. Sp. Sci., vol. 28, no. 1, pp. 34–52, 2025, doi: 10.1016/j.ejrs.2024.12.001.
N. A. Sameer and B. M. Nema, “LSTM and CNN hybrid model for enhanced fingerprint recognition,” Emerg. Trends Drugs, Addict. Heal., vol. 5, no. January, p. 100174, 2025, doi: 10.1016/j.etdah.2025.100174.
X. Zeng, N. Abdullah, and B. Liang, “A widely applicable and robust LightGBM - Artificial neural network forecasting model for short-term wind power density,” Heliyon, vol. 9, no. 12, p. e23071, 2023, doi: 10.1016/j.heliyon.2023.e23071.
Y. Song, S. Kuang, J. Huang, and D. Zhang, “Unsupervised anomaly detection of industrial building energy consumption,” Energy Built Environ., no. November 2024, 2025, doi: 10.1016/j.enbenv.2024.12.005.
E. Chandralekha, S. Vinodhini, V. Kandasamy, and P. Rama, “ScienceDirect ScienceDirect Heart Rate Rate Anomaly Anomaly Detection Detection in in Healthcare Healthcare Using Using Elliptic Elliptic Envelope and Local Forest Envelope and Local Forest,” Procedia Comput. Sci., vol. 258, pp. 1677–1687, 2025, doi: 10.1016/j.procs.2025.04.402.
D. Jiandong, W. Peng, M. Wentao, F. Shuai, and H. Zequan, “International Journal of Electrical Power and Energy Systems A novel hybrid model based on nonlinear weighted combination for short-term wind power forecasting ☆,” Int. J. Electr. Power Energy Syst., vol. 134, no. December 2020, p. 107452, 2022, [Online]. Available: https://doi.org/10.1016/j.ijepes.2021.107452
A. Bonfigli, L. Bacco, L. Pecchia, M. Merone, and F. Dell’Orletta, “Efficient multi-task learning with instance selection for biomedical NLP,” Comput. Biol. Med., vol. 190, no. August 2024, p. 110050, 2025, doi: 10.1016/j.compbiomed.2025.110050.
C. Zhang et al., “Exploring Psychological Trends in Populations With Chronic Obstructive Pulmonary Disease During COVID-19 and Beyond: Large-Scale Longitudinal Twitter Mining Study,” J. Med. Internet Res., vol. 27, 2025, doi: 10.2196/54543.
S. V. Oprea and A. Bâra, “Unveiling the nexus between energy storage and electricity markets in academic publications. A data-driven analysis of emerging trends and market dynamics using NLP, sentiment analysis and probabilistic modeling,” J. Energy Storage, vol. 106, no. August 2024, 2025, doi: 10.1016/j.est.2024.114917.
M. Mohandes, A. Hussain, P. Khan, and H. Nuha, “Non-Vital Privacy Compromise for Improved Services in Pilgrimage Seasons,” 2019.
DOI: http://dx.doi.org/10.26798/jiko.v9i3.2163
Article Metrics
Abstract view : 0 timesPDF (Bahasa Indonesia) - 0 times
Refbacks
- There are currently no refbacks.
Copyright (c) 2025 Listiana Dewi Milasari
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.