A Literature Review of Quality Evaluation of Large-Scale Recommendation Systems Techniques

The sudden increase of online or internet based cooperates have led to the migration to RS. These systems shall provide accurate predictions and recommendations of services and products to the users of the same interest. The recommendation and prediction performance should strictly audited and evaluated to maintain the optimum quality of service that is being served and to ensure the continuity of such technologies through various considered factors. However, due to the exponential growth of number of the services available online, new challenges have erupted leading to many defects that affect drastically the quality of accurate prediction and recommendation of these systems such as data sparsity, the problem of scalability and cold start. These challenges have attracted many researchers and data scientists to investigate and further exploration of the main source of these raising issues especially in large scale and distributed systems.