Penerapan Algoritma Harmony Search untuk Optimasi Fuzzy C-Means dalam Sistem Rekomendasi Film Berbasis Hybrid Filtering

Muftah, Hi M Naser (2025) Penerapan Algoritma Harmony Search untuk Optimasi Fuzzy C-Means dalam Sistem Rekomendasi Film Berbasis Hybrid Filtering. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) COVER.pdf Download (5MB)
	Text (Bab 1) BAB 1.pdf Download (225kB)
	Text (Bab 2) BAB 2.pdf Restricted to Repository staff only until 4 December 2028. Download (415kB)
	Text (Bab 3) BAB 3.pdf Restricted to Repository staff only until 4 December 2028. Download (528kB)
	Text (Bab 4) BAB 4.pdf Restricted to Repository staff only until 4 December 2028. Download (1MB)
	Text (Bab 5) BAB 5.pdf Download (213kB)
	Text (Daftar Pustaka) DAFTAR PUSTAKA.pdf Download (201kB)
	Text (Lampiran) LAMPIRAN.pdf Restricted to Repository staff only Download (241kB)

Abstract

The rapid growth of movie data has introduced significant challenges in recommendation systems, particularly related to cold start and sparsity issues. This research aims to develop a movie recommendation model based on Hybrid Filtering by integrating the Harmony Search algorithm for optimizing initial centroids, Fuzzy C-Means as the user clustering method, and a weighted combination of Collaborative Filtering and ContentBased Filtering using an alpha parameter to produce more accurate recommendations. The dataset used in this study is MovieLens 1M, which has undergone several preprocessing stages. The overall process includes constructing user profile vectors, optimizing centroids using Harmony Search, forming clusters with Fuzzy C-Means, calculating recommendation scores based on user similarity and content similarity, and implementing the model into an API and website for real-world testing. Evaluation results show that the best configuration is obtained with 4 clusters, a fuzzification value of m = 1.5, and an alpha value of 0.7, achieving an RMSE of 0.897442, MAE of 0.701134, Precision of 0.762984, and Recall of 0.547199. This performance surpasses the use of single-method approaches, where Content-Based Filtering produced significantly lower metrics and pure Collaborative Filtering exhibited signs of overfitting. These findings demonstrate that integrating Harmony Search, Fuzzy CMeans, and Hybrid Filtering effectively addresses sparsity and cold start issues, resulting in more relevant movie recommendations.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Eka, Prakarsa Mandyartha	NIDN0725058805	eka_prakarsa.fik@upnjatim.ac.id
Thesis advisor	Muhammad Muharrom, Al Haromainy	NIDN0701069503	muhammad.muharrom.if@upnjatim.ac.id

Subjects:

T Technology > T Technology (General)

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Muftah Hi M Naser

Date Deposited:

04 Dec 2025 07:38

Last Modified:

04 Dec 2025 07:38

URI:

https://repository.upnjatim.ac.id/id/eprint/47856