This study assesses the performance of selected classifiers concerning the overall performance of the selected classifiers and proposes an anime tv series based on genre using two different techniques of machine learning (ML). In the study, the classifiers, Random Forest and Decision Tree were used. The procedure has three steps which are data selection, training the classifier, and evaluation of the classifier. Of the total dataset, 70% was used to train the ML classifier and 30% of the dataset was used to test it. In SPSS, measures alpha and CI were 0.95 and 0.03, respectively. According to an experimental investigation by using Google Colab, two classifiers were evaluated against a dataset with a random selection of animes deciding which ones were addictive. As for accuracy level, SVD outperformed Random Forest; by obtaining a score of 94.73% for amnasty_cf’s Random Forest and 93.84% for the Decision Tree Classifier. It is shown that the Random Forest approach considerably improves the satisfaction of the anime recommendation compared to the Decision Tree one. Random Forest is seen as an extension of Decision Trees where each Decision Tree built to cooperate according to the user-specified genre choices and provides more precise, personalised recommendations. This technique is superior to Decision Trees’ single-tree method that underscores the choice of algorithms in recommendation systems. It is recommended that Random Forest be used to improve user experience on the anime recommendations.