Recommendation systems are designed to investigate and suggest relevant items based on the user-s preferences. Recommender systems are the fundamental tool in informed consumption, services, and decision-making for effective recommendations. One of the common challenges faced by recommendation systems is data sparsity which leads the system to decrease prediction efficiency. Data sparsity refers to the difficulty in discovering enough reliable data in the system. Generally used approaches to solve data sparsity issues are collaborative filtering and content-based filtering. However, such approaches are inadequate to alleviate the data sparsity. Transfer Learning is the learning paradigm that utilizes learned knowledge from one task in a different but related source domain to solve the task in the other target domain. Transfer learning supports the recommender system by learning knowledge from various users based on their similarities to produce useful suggestions. Application of transfer learning in recommender system properly reduces the data sparsity problem and increases the efficiency of the recommender system.