Benchmarking for Recommender System (MFRISE)

The advent of the internet age offers overwhelming choices movies and shows to viewers which create need comprehensive Recommendation Systems (RS). System will suggest best content based on their choice using methods Information Retrieval, Data Mining Machine Learning algorithms. novel Multifaceted Engine (MFRISE) algorithm proposed in this paper help users get personalized movie recommendations multi-clustering approach user cluster Movie along with interaction effect. This add value our existing parameters like ratings reviews. In real-world scenarios, recommenders have many non-functional requirements technical nature. Evaluation must take these issues into account order produce good recommendations. show various evaluation RMSE, MAE timings, can be used measure accuracy speed Recommender system. benchmarking results also helpful for new recommendation has MovieLens dataset purpose experimentation. studied consider both quantitative qualitative aspects mean squared error (MSE), root (RMSE), Test Time Fit are calculated each popular recommender (NMF, SVD, SVD++, SlopeOne, Co- Clustering) implementation. study identifies gaps challenges faced by above algorithm. researchers propose algorithms overcoming identified research