Anime Recommendation System (ARS) presents a creative solution to guide the newly entered candidates and participants alike in exploring the vast and diverse anime landscape. By manipulating a hybrid approach that combines the content-based filtering, collaborative filtering with the technique of machine learning (ML) like the k-nearest neighbor (KNN) algorithm, Logistic Regression (LR), Multinomial NB (MNB) and Random Forest (RF) and ARS flourish the personalized and context-aware anime recommendations. The system uses to be compatible in adapting dynamically to user preferences choices and feedback, making the recommendations user friendly, relevant and developing gradually with time. This paper contributes to the development, evaluation and implementation of ARS, in compare to it with current existing anime recommendation systems. The output depicts the ARS to be highly efficient and accurate in identifying anime titles put in order with users’ preferences using the Random Forest (RF) algorithm achieved with a highest accuracy at 75.98%, followed by Logistic Regression (LR) with 74.95%, KNN with 73.68%, and Multinomial NB with 72.18%.