Please use this identifier to cite or link to this item:
https://repositori.uma.ac.id/handle/123456789/22762
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Muliono, Rizki | - |
dc.contributor.author | Sitorus, Muhammad Rizky | - |
dc.date.accessioned | 2024-01-23T03:13:25Z | - |
dc.date.available | 2024-01-23T03:13:25Z | - |
dc.date.issued | 2023-09-14 | - |
dc.identifier.uri | https://repositori.uma.ac.id/handle/123456789/22762 | - |
dc.description | 56 Halaman | en_US |
dc.description.abstract | Sistem rekomendasi film telah menjadi elemen penting dalam berbagai aplikasi dan situs web di era digital saat ini. Sistem ini membantu pengguna menemukan film-film yang sesuai dengan minat dan preferensi mereka. Terdapat berbagai jenis sistem rekomendasi seperti Content Based Filtering, Collaborative Filtering dan Hybrid Filtering (gabungan dari Content Based Filtering dan Collaborative Filtering). Dalam penelitian ini, dilakukan implementasi Model-Based Collaborative Filtering menggunakan algoritma Singular Value Decomposition (SVD) pada sistem rekomendasi film. Algoritma SVD digunakan untuk menganalisis interaksi pengguna dengan film-film tersebut dan menghasilkan representasi latent yang lebih ringkas tetapi tetap informatif. Dengan demikian, rekomendasi film dapat dibuat berdasarkan preferensi yang mendasari perilaku pengguna. Hasil evaluasi prediksi menunjukkan bahwa penggunaan algoritma SVD dengan nilai k = 3000 dalam sistem rekomendasi film menghasilkan nilai evaluasi RMSE (Root Mean Square Error) sebesar 0,4002 yang menunjukkan tingkat akurasi model yang baik. Selain itu, nilai Mean Absolute Error (MAE) yang dihasilkan adalah 0,1186 yang menunjukkan tingkat kesalahan prediksi yang rendah. Film recommendation systems have become a crucial element in various applications and websites in the current digital era. These systems assist users in discovering films that align with their interests and preferences. There are various types of recommendation systems such as Content-Based Filtering, Collaborative Filtering, and Hybrid Filtering (a combination of Content-Based Filtering and Collaborative Filtering). In this research, the implementation of Model-Based Collaborative Filtering using the Singular Value Decomposition (SVD) algorithm in the film recommendation system is carried out. The SVD algorithm is utilized to analyze user interactions with these films and generate more concise yet informative latent representations. Therefore, film recommendations can be made based on the underlying user preferences. Prediction evaluation results indicate that the use of the SVD algorithm with a value of k = 3000 in the film recommendation system yields an RMSE (Root Mean Square Error) evaluation score of 0.4002, demonstrating a high level of model accuracy. Furthermore, the Mean Absolute Error (MAE) generated is 0.1186, indicating a low level of prediction error. | en_US |
dc.language.iso | id | en_US |
dc.publisher | Universitas Medan Area | en_US |
dc.relation.ispartofseries | NPM;188160044 | - |
dc.subject | model-based | en_US |
dc.subject | collaborative filtering | en_US |
dc.subject | singular value decomposition (svd) | en_US |
dc.subject | sistem rekomendasi film | en_US |
dc.subject | rating | en_US |
dc.subject | film recommendation system | en_US |
dc.title | Implementasi Model-Based Collaborative Filtering pada Sistem Rekomendasi Film Menggunakan Algoritma SVD (Singular Value Decomposition) | en_US |
dc.title.alternative | Implementation of Model-Based Collaborative Filtering in a Film Recommendation System Using the SVD (Singular Value Decomposition) Algorithm | en_US |
dc.type | Skripsi Sarjana | en_US |
Appears in Collections: | SP - Informatic Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
188160044 - Muhammad Rizky Sitorus - Fulltext.pdf | Cover, Abstract, Chapter I, II, III, V, Bibliography | 830.1 kB | Adobe PDF | View/Open |
188160044 - Muhammad Rizky Sitorus - Chapter IV.pdf Restricted Access | Chapter IV | 612.93 kB | Adobe PDF | View/Open Request a copy |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.