Sistem Rekomendasi Destinasi Wisata Berbasis Pemeringkatan Rating, Preferensi, dan Pola Perjalanan Wisatawan Domestik Menggunakan Multi-Model Collaborative Filtering

Marzuki, Imam (2025) Sistem Rekomendasi Destinasi Wisata Berbasis Pemeringkatan Rating, Preferensi, dan Pola Perjalanan Wisatawan Domestik Menggunakan Multi-Model Collaborative Filtering. Doctoral thesis, Institut Teknologi Sepuluh Nopember.

Text 7022222004-Dotoral.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

Penelitian ini mengembangkan sistem rekomendasi destinasi wisata menggunakan Multi-Model Collaborative Filtering untuk mengatasi rendahnya relevansi rekomendasi dan permasalahan cold start pada platform pariwisata digital. Penelitian ini menggunakan tiga pendekatan machine learning yang meliputi model Neural Collaborative Filtering untuk memprediksi rating, model KNN-Based Stacking untuk mengelompokkan preferensi user, dan metode Hierarchical Clustering berbasis Cosine Distance untuk memetakan pola perjalanan wisatawan. Penelitian ini menguji ketiga model tersebut pada dataset yang berisi data demografi, riwayat perjalanan, dan rating destinasi wisata untuk mengevaluasi kemampuan model dalam memahami perilaku user. Penelitian ini menghasilkan peningkatan akurasi prediksi rating, peningkatan ketepatan klasifikasi preferensi, serta pemetaan pola perjalanan yang konsisten melalui integrasi ketiga metode tersebut. Penelitian ini menyimpulkan bahwa pendekatan multi-model mampu menghasilkan rekomendasi yang lebih relevan, adaptif, dan holistik bagi user dalam konteks pengembangan sistem rekomendasi pariwisata berbasis data.
===================================================================================================================================
This study develops a tourism destination Recommendation system using a Multi-Model Collaborative Filtering approach to address the issues of low Recommendation relevance and cold-start problems commonly found in digital tourism platforms. The study employs three machine learning methods, including a Neural Collaborative Filtering model to predict user ratings, a KNN-Based Stacking model to classify user preferences, and a Hierarchical Clustering method using Cosine Distance to map tourist travel patterns. The study evaluates these models using a dataset containing demographic information, travel histories, and destination ratings to assess their capability in capturing user behavior. The study demonstrates that the integrated models improve rating prediction accuracy, enhance preference classification performance, and produce consistent travel-pattern groupings. The study concludes that the multi-model approach provIDes more relevant, adaptive, and holistic Recommendations for users, thereby supporting the development of data-driven tourism Recommendation systems.

Item Type:	Thesis (Doctoral)
Uncontrolled Keywords:	Sistem Rekomendasi, Multi-Model Collaborative Filtering , Neural Collaborative Filtering (NCF), KNN-Based Stacking, Hierarchical Clustering, Recommender System, Multi-Model Collaborative Filtering , Neural Collaborative Filtering (NCF), KNN-Based Stacking, Hierarchical Clustering
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20001-(S3) PhD Thesis
Depositing User:	Imam Marzuki
Date Deposited:	07 Jan 2026 03:41
Last Modified:	07 Jan 2026 03:41
URI:	http://repository.its.ac.id/id/eprint/129322

Actions (login required)

View Item