A grasshopper optimization algorithm-based movie recommender system

Abstract

A movie recommendation system functions as a specialized information system, providing users with personalized suggestions aligned with their movie preferences. Employing advanced algorithms and data analysis methods, these systems scrutinize variables such as users' viewing history and preferences to formulate personalized recommendations. Our proposed methodology, termed GOA-k-means, amalgamates the Grasshopper Optimization Algorithm (GOA) with k-means clustering to navigate the dynamic nature of user preferences. Facilitating real-time calibration, GOA-k-means yields recommendations that adapt to users' shifting interests. We developed our model utilizing a dataset of one million records from Movielens, pre-processed via z-score normalization and subjected to Principal Component Analysis (PCA) for feature extraction. In comparison to conventional techniques, GOA-k-means demonstrated superior performance in metrics such as precision, recall, Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE), establishing itself as a valuable tool for augmenting user engagement in the entertainment industry. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2023.