Search engine for discovering works of Art, research articles, and books related to Art and Culture
ShareThis
Javascript must be enabled to continue!

Film Recommender System Menggunakan Metode Neural Collaborative Filtering

View through CrossRef
Pada saat ini media hiburan telah berkembang pesat dan tersedia secara digital. Hiburan khususnya dalam bentuk film semakin tersedia secara luas. Keinginan untuk menikmati hiburan dalam media digital mendorong pengguna internet lain untuk mengunjungi situs-situs yang menawarkan film tertentu, sehingga meningkatkan minat mereka terhadap website yang menawarkan hiburan digital. Tidak semua situs penyedia hiburan digital menyajikan item yang menjanjikan kepuasan pengguna. Sebuah item yang sama tidak tentu akan disukai oleh semua user dan terbatasnya informasi yang disediakan menjadi salah satu kendala bagi pengguna sehingga membutuhkan waktu untuk pengguna menemukan film yang sesuai. Oleh karena itu recommender system dibutuhkan dalam memberikan informasi berdasarkan kebutuhan pengguna. Recommender system akan membantu seorang user dalam mencari sebuah item yang berdasarkan ketertarikan masing-masing dengan memberikan prediksi beberapa item berdasarkan preferensi user yang berasal dari riwayat penilaian user terhadap item tersebut. Recommender system juga telah mengalami kemajuan dalam mengimplementasikan metode. Deep learning yang merupakan salah satu penemuan dalam metode recommender system dirancang untuk mengatasi beberapa kekurangan dari teknlogi lain dan memberikan revolusi arsitektur rekomendasi dalam meningkatkan kinerja dalam pemberian prediksi. Penelitian ini menggunakan pendekatan prediksi Collaborative Filtering dengan mengimplementasikan deep learning berdasarkan teknologi Neural Collaborative Filtering pada dataset MovieLens. Evaluasi model dilakukan menggunakan metrik skor regresi Root Mean Square Error (RMSE). Hasil pada pengujian model menunjukkan hasil terbaik dengan nilai rata-rata loss value sebesar 0,1356 pada fase train dan sebesar 0,8898 pada fase val, dengan learning rate dan batch size memperoleh kinerja terbaik ketika learning rate bernilai 0,001 dan batch size dengan nilai 1024. Abstract At this time entertainment media has become available digitally. Entertainment especially in the form of movies is increasingly widely available. The desire to enjoy entertainment in digital media encourages other internet users to visit sites that offer certain movies, thus increasing interest in websites that offer digital entertainment. Not all digital entertainment provider sites present items that promise user satisfaction. The same item will not necessarily liked by all users and the limited information is one of the obstacles for users so that it takes time for users to find the right film. Therefore, a recommendation system is needed in providing information based on user needs. The recommendation system will help users find items based on their respective interests by providing predictions. The recommender system will help a user find an item based on their respective interests by providing predictions of several items based on user preferences derived from the user's assessment history of the item. The recommendation system has also made progress in implementing the method. Deep learning which is one of the discoveries in the recommender system method is designed to overcome some of the shortcomings of other technologies and provide a recommendation architecture revolution in improving performance in delivery. This study using a Collaborative Filtering prediction approach by implementing deep learning based on Neural Collaborative Filtering technology on the MovieLens dataset. The evaluation of the model was carried out using the Root Mean Square Error regression score metric. The results on the model test show the best results with can average loss value of 0,1356 on the train label and 0,8898 on the val label, with the learning rate and batch size getting the best performance when the learning rate is 0,001 and the batch size is 1024.
Title: Film Recommender System Menggunakan Metode Neural Collaborative Filtering
Description:
Pada saat ini media hiburan telah berkembang pesat dan tersedia secara digital.
Hiburan khususnya dalam bentuk film semakin tersedia secara luas.
Keinginan untuk menikmati hiburan dalam media digital mendorong pengguna internet lain untuk mengunjungi situs-situs yang menawarkan film tertentu, sehingga meningkatkan minat mereka terhadap website yang menawarkan hiburan digital.
Tidak semua situs penyedia hiburan digital menyajikan item yang menjanjikan kepuasan pengguna.
Sebuah item yang sama tidak tentu akan disukai oleh semua user dan terbatasnya informasi yang disediakan menjadi salah satu kendala bagi pengguna sehingga membutuhkan waktu untuk pengguna menemukan film yang sesuai.
Oleh karena itu recommender system dibutuhkan dalam memberikan informasi berdasarkan kebutuhan pengguna.
Recommender system akan membantu seorang user dalam mencari sebuah item yang berdasarkan ketertarikan masing-masing dengan memberikan prediksi beberapa item berdasarkan preferensi user yang berasal dari riwayat penilaian user terhadap item tersebut.
Recommender system juga telah mengalami kemajuan dalam mengimplementasikan metode.
Deep learning yang merupakan salah satu penemuan dalam metode recommender system dirancang untuk mengatasi beberapa kekurangan dari teknlogi lain dan memberikan revolusi arsitektur rekomendasi dalam meningkatkan kinerja dalam pemberian prediksi.
Penelitian ini menggunakan pendekatan prediksi Collaborative Filtering dengan mengimplementasikan deep learning berdasarkan teknologi Neural Collaborative Filtering pada dataset MovieLens.
Evaluasi model dilakukan menggunakan metrik skor regresi Root Mean Square Error (RMSE).
Hasil pada pengujian model menunjukkan hasil terbaik dengan nilai rata-rata loss value sebesar 0,1356 pada fase train dan sebesar 0,8898 pada fase val, dengan learning rate dan batch size memperoleh kinerja terbaik ketika learning rate bernilai 0,001 dan batch size dengan nilai 1024.
 Abstract At this time entertainment media has become available digitally.
Entertainment especially in the form of movies is increasingly widely available.
The desire to enjoy entertainment in digital media encourages other internet users to visit sites that offer certain movies, thus increasing interest in websites that offer digital entertainment.
Not all digital entertainment provider sites present items that promise user satisfaction.
The same item will not necessarily liked by all users and the limited information is one of the obstacles for users so that it takes time for users to find the right film.
Therefore, a recommendation system is needed in providing information based on user needs.
The recommendation system will help users find items based on their respective interests by providing predictions.
The recommender system will help a user find an item based on their respective interests by providing predictions of several items based on user preferences derived from the user's assessment history of the item.
The recommendation system has also made progress in implementing the method.
Deep learning which is one of the discoveries in the recommender system method is designed to overcome some of the shortcomings of other technologies and provide a recommendation architecture revolution in improving performance in delivery.
This study using a Collaborative Filtering prediction approach by implementing deep learning based on Neural Collaborative Filtering technology on the MovieLens dataset.
The evaluation of the model was carried out using the Root Mean Square Error regression score metric.
The results on the model test show the best results with can average loss value of 0,1356 on the train label and 0,8898 on the val label, with the learning rate and batch size getting the best performance when the learning rate is 0,001 and the batch size is 1024.

Related Results

Enhanced Product Review Recommendations Using Collaborative Filtering and Singular Value Decomposition
Enhanced Product Review Recommendations Using Collaborative Filtering and Singular Value Decomposition
Recommender systems have become indispensable tools for enhancing user satisfaction and engagement across diverse business sectors, including online marketplaces, streaming service...
Film Recommender System Menggunakan Metode Neural Collaborative Filtering
Film Recommender System Menggunakan Metode Neural Collaborative Filtering
Pada saat ini media hiburan telah berkembang pesat dan tersedia secara digital. Hiburan khususnya dalam bentuk film semakin tersedia secara luas. Keinginan untuk menikmati hiburan ...
Reclaiming the Wasteland: Samson and Delilah and the Historical Perception and Construction of Indigenous Knowledges in Australian Cinema
Reclaiming the Wasteland: Samson and Delilah and the Historical Perception and Construction of Indigenous Knowledges in Australian Cinema
It was always based on a teenage love story between the two kids. One is a sniffer and one is not. It was designed for Central Australia because we do write these kids off there. N...
EVALUATION OF HYBRID MOVIE RECOMMENDATION SYSTEM BASED ON NEURAL NETWORKS
EVALUATION OF HYBRID MOVIE RECOMMENDATION SYSTEM BASED ON NEURAL NETWORKS
Abstract: Recommendation systems are becoming increasingly important with the growth of streaming platforms. The purpose of this study is to compare the performance of Content-Base...
Alternative Entrances: Phillip Noyce and Sydney’s Counterculture
Alternative Entrances: Phillip Noyce and Sydney’s Counterculture
Phillip Noyce is one of Australia’s most prominent film makers—a successful feature film director with both iconic Australian narratives and many a Hollywood blockbuster under his ...
Development of E-Commerce Website Recommender System Using Collaborative Filtering and Deep Learning Techniques
Development of E-Commerce Website Recommender System Using Collaborative Filtering and Deep Learning Techniques
Recommender system or recommendation system is becoming an increasingly important technology on e-commerce websites to help users find products that suit their preferences. However...
A Recommender System for Personalized Reading Recommendations and Literature Discovery utilizing the HGRNN-EOO technique
A Recommender System for Personalized Reading Recommendations and Literature Discovery utilizing the HGRNN-EOO technique
Recommender systems are used to address information overload, enhance personalization, and improve user experience by providing tailored suggestions based on individual preferences...
Cinema as a form of composition
Cinema as a form of composition
Technique and creativity Having been called upon to provide a contribution to a publication dedicated to “Techne”, I feel it is fitting to start from the theme of technique, ...

Back to Top