Search for collections on Undip Repository

REKOMENDASI DESTINASI PENYANGGA DI KAWASAN BOROBUDUR MENGGUNAKAN MACHINE LEARNING BERBASIS TOURIST PERCEPTION, SOCIAL, AND DEMOGRAPHIC MODEL

AGUSTINA, Candra and Purwanto, Purwanto and Farikhin, Farikhin (2025) REKOMENDASI DESTINASI PENYANGGA DI KAWASAN BOROBUDUR MENGGUNAKAN MACHINE LEARNING BERBASIS TOURIST PERCEPTION, SOCIAL, AND DEMOGRAPHIC MODEL. Doctoral thesis, UNIVERSITAS DIPONEGORO.

[thumbnail of Awal-1.pdf] Text
Awal-1.pdf

Download (349kB)
[thumbnail of Awal.pdf] Text
Awal.pdf
Restricted to Repository staff only

Download (1MB)
[thumbnail of BAB I.pdf] Text
BAB I.pdf

Download (334kB)
[thumbnail of BAB II.pdf] Text
BAB II.pdf

Download (796kB)
[thumbnail of BAB III.pdf] Text
BAB III.pdf
Restricted to Repository staff only

Download (1MB)
[thumbnail of BAB IV.pdf] Text
BAB IV.pdf
Restricted to Repository staff only

Download (1MB)
[thumbnail of BAB V.pdf] Text
BAB V.pdf
Restricted to Repository staff only

Download (344kB)
[thumbnail of Daftar Pustaka.pdf] Text
Daftar Pustaka.pdf

Download (432kB)
[thumbnail of Lampiran.pdf] Text
Lampiran.pdf
Restricted to Repository staff only

Download (9MB)

Abstract

Wisatawan seringkali menghadapi kesulitan dalam menentukan destinasi selanjutnya yang sesuai dengan minat mereka, sehingga diperlukan sebuah sistem yang dapat membantu memilih destinasi dengan tingkat kepuasan yang tinggi. Latar belakang penelitian ini adalah pertumbuhan pariwisata yang pesat dan kemunculan berbagai destinasi wisata baru di sekitar wilayah Borobudur dan Yogyakarta. Penelitian ini menggunakan beberapa tahapan metodologi utama. Pertama, analisis sentimen pengunjung Candi Borobudur dilakukan menggunakan algoritme Naive Bayes dan teknik SMOTE untuk menangani ketidakseimbangan data, kemudian hasilnya dimasukkan ke dalam dataset. Kedua, destinasi wisata dikelompokkan menggunakan algoritme K-means, yang menghasilkan dua kelompok yang dijadikan label dalam dataset. Ketiga, pengolahan dataset dilakukan menggunakan beberapa algoritme klasifikasi seperti Decision Tree (DT), Naive Bayes (NB), dan K-Nearest Neighbors (KNN), dengan seleksi fitur menggunakan Forward Selection.
Hasil penelitian menunjukkan bahwa Algoritme terbaik untuk memprediksi destinasi wisata adalah Decision Tree dengan fitur seleksi forward selection, yang mencapai tingkat akurasi sebesar 81,87%. Fitur-fitur yang terpilih dalam model ini meliputi persepsi wisatawan tentang fasilitas, kualitas promosi, dan penggunaan media sosial untuk mendapatkan ide. Selain itu, demografi wisatawan yang digunakan adalah status tempat tinggal wisatawan, apakah milik sendiri atau sewa. Model yang dikembangkan mampu memberikan rekomendasi destinasi wisata dengan tingkat akurasi yang tinggi, sehingga diharapkan dapat memberikan manfaat bagi wisatawan dan pelaku usaha wisata dalam meningkatkan kepuasan dan pengalaman wisata.
Kata kunci: rekomendasi destinasi wisata, machine learning, analisis sentimen, Decision Tree, forward selection, Borobudur

Tourist often encounter difficulties in determining an appropriate destination that aligns with their interests, necessitating a system capable of selecting destinations with high satisfaction levels. The background of this study is the rapid growth of tourism and the emergence of various new tourist destinations in the surrounding areas of Borobudur and Yogyakarta. This research employs several main methodological phases. First, sentiment analysis of Candi Borobudur visitors is conducted using the Naive Bayes algorithm and SMOTE technique to handle data imbalance, with the results incorporated into the dataset. Second, tourist destinations are grouped using the K-means algorithm, resulting in two labeled groups in the dataset. Third, data processing is performed using several classification algorithms such as Decision Tree (DT), Naive Bayes (NB), and K-Nearest Neighbors (KNN), with feature selection using Forward Selection. The results show that the best algorithm for predicting tourist destinations is the Decision Tree with forward feature selection, achieving an accuracy of 81.87%. The selected features in this model include tourists' perceptions of facilities, promotional quality, and the use of social media for ideas. Additionally, demographic factors used include the status of the tourist's place of residence and whether they own or rent the property. The developed model is capable of providing destination recommendations with high accuracy, which is expected to benefit both tourists and tourism industry stakeholders in improving satisfaction and travel experiences.
Keywords: destination recommendations, machine learning, sentiment analysis, Decision Tree, forward selection, Borobudur.

Item Type: Thesis (Doctoral)
Uncontrolled Keywords: rekomendasi destinasi wisata, machine learning, analisis sentimen, Decision Tree, forward selection, Borobudur
Subjects: Sciences and Mathemathic
Divisions: Postgraduate Program > Doctor Program in Information System
Depositing User: ekana listianawati
Date Deposited: 02 Jul 2025 04:37
Last Modified: 02 Jul 2025 04:37
URI: https://eprints2.undip.ac.id/id/eprint/34173

Actions (login required)

View Item View Item