Finding Depth-k Skyline Friends in Distributed Social Networks by Using Neuro-fuzzy Networks

Analyzing the information on social networks and identifying similar users are crucial functions of recommendation systems. In recent years, skyline queries have been applied to search for similar users and aided by neural networks to speed up the search. However, most of these methods can only be used in centralized environments and not on distributed environments, which most social networks are operated in. Previous algorithms have also been limited by the use of neural networks, which made searches slower. This study proposed an algorithm that can search for similar users in distributed systems and is accelerated using back-propagation neuro-fuzzy networks. Simulation results demonstrate the efficacy and efficiency of the proposed approach.