Genre-Based Movie Recommender System with XGBoost
Suresh M. Kumar, Jyoti Prakash Singh, Shantanu Shantanu · 2024
Recommender systems are crucial components of online shopping platforms, as they enhance sales by suggesting relevant products to users. It provides personalized recommendations of products or items depending on the user's preferences, effectively addressing the problem of information overload. The recommender system is an information filtering method that effectively sifts through large volumes of data. Algorithms such as Matrix factorization, K-nearest neighbours, and Convolutional neural networks have already been employed to generate suggestions. This study presents a formulation of the recommendation issue as a supervised model. We employ the extreme gradient boosting (XGBoost) method to predict the rating score of movies and recommend the highest-rated movie to the user. The model underwent training and testing using the Movilens dataset, surpassing the performance of existing models in terms of precision and recall. It achieved a precision of 0.89, a recall of 0.63, an F-1 score of 0.74, an RMSE of 0.97, and an MAE of 0.78.