OPTIMASI SVR UNTUK PREDIKSI RATING FILM BERBASIS METADATA TMDB

Sendi Ade Viransyah; Ade Irma Purnamasari; Denni Pratama; Edi Wahyudin; Edi Tohidi

Authors

Sendi Ade Viransyah STMIK IKMI Cirebon, Indonesia
Ade Irma Purnamasari STMIK IKMI Cirebon, Indonesia
Denni Pratama STMIK IKMI Cirebon, Indonesia
Edi Wahyudin STMIK IKMI Cirebon, Indonesia
Edi Tohidi STMIK IKMI Cirebon, Indonesia

Keywords:

feature scaling, hyperparameter tuning, kernel optimization

Abstract

Prediksi rating film merupakan kebutuhan penting dalam industri hiburan untuk mendukung pengambilan keputusan sebelum perilisan film. Namun, kompleksitas metadata film dengan rentang nilai fitur yang berbeda serta hubungan non-linear antar variabel menyebabkan model prediksi sering menghasilkan akurasi yang rendah. Penelitian ini bertujuan meningkatkan akurasi prediksi rating film menggunakan algoritma Support Vector Regression melalui pendekatan tiga tahap optimasi, yaitu feature scaling, pemilihan kernel, dan penyesuaian hyperparameter. Dataset yang digunakan adalah TMDB Movie Dataset dengan sepuluh fitur numerik yang telah melalui tahap preprocessing dan seleksi fitur. Tiga metode scaling diuji, empat jenis kernel dibandingkan, dan hyperparameter dioptimasi menggunakan GridSearchCV. Hasil penelitian menunjukkan bahwa kombinasi MinMaxScaler, kernel RBF, serta parameter optimal menghasilkan performa terbaik dan mampu menurunkan nilai error secara signifikan dibandingkan konfigurasi baseline serta model pembanding lainnya. Temuan ini menunjukkan bahwa strategi optimasi terintegrasi pada SVR efektif dalam meningkatkan akurasi prediksi rating film dan relevan untuk analisis performa film sebelum rilis.

References

H. Mahmud, et al., A Machine Learning Approach to Predict Movie Revenue Based on Pre-Released Metadata, Journal of Computer Science, 2020.

F. Nurhalizah, et al., Supervised Learning for Movie Rating Prediction, Journal of Data Science, 2024.

X. Zhang and Q. Bai, Prediction of Movie Popularity Using Supervised Learning Techniques, Applied Computing Review, 2023.

J. Permana and A. T. Wibowo, Content-Based Movie Recommendation Using TMDB Metadata, IJoICT, 2023.

T. A. Toha, et al., Effect of Feature Scaling on Regression Stability, Data Science Review, 2022.

A. Pinheiro, et al., The Impact of Feature Scaling on Machine Learning Algorithms, arXiv preprint, 2025.

M. Fan, et al., The Role of Preprocessing in Large-Scale Machine Learning, Journal of Big Data, 2021.

J. Li, et al., Dataset Quality and Its Influence on Regression Performance, Information Processing Letters, 2021.

M. R. Apriyadi, Ermatita, and D. P. Rini, Hyperparameter Optimization of Support Vector Regression Algorithm Using Metaheuristic Algorithms, IJACSA, vol. 14, no. 2, 2023.

H. Azis, et al., Performance Analysis of Support Vector Regression in Predicting National Commodity Prices, ILKOM Jurnal Ilmiah, vol. 15, no. 2, 2023.

R. Ishlah, M. Jamal, and I. Fatkhurrohman, Implementation of GridSearchCV on SVR for Stock Prediction, Jurnal Informatika, 2023.

S. Novianto, et al., Forecasting Energy Consumption Using Support Vector Regression, JIPI, 2024.

F. Fajri and H. Primajaya, Comparison Between Grid Search and Random Search for Hyperparameter Optimization, JAIC, 2023.

R. Rusman, et al., Genetic Algorithm Optimization for SVR Performance, ICIC Express Letters, 2023.

Z. Zhang and M. Bai, Advances in Movie Popularity Forecasting Using Regression Models, Entertainment Computing, 2023.

L. T. Emiyati, et al., Hybrid Machine Learning Framework for Regression Analysis, International Journal of Computer Applications, 2022.

K. Putra and S. Adinata, Scaling Strategies for Regression Models Using Sparse Data, Journal of Intelligent Systems, 2021.

D. Wicaksana, et al., Optimization of Kernel Functions in Support Vector Regression, Electronics and Informatics Journal, 2023.

A. Lestari and H. Prasetyo, Hyperparameter Optimization Approaches for Non-Linear Regression Models, International Journal of Artificial Intelligence, 2024.

B. Andriansyah, et al., Comparative Study of Regression Algorithms on Non-Linear Datasets, Journal of Information Technology Research, 2022.

R. Saputra and N. Syahputra, Evaluation Metrics for Regression-Based Prediction Models, Data Science Review, 2023.