Hybrid Privacy for Social Media Content using Convolutional Neural Networks (CNNs)

Abstract

One of the fast-growing fields in computer science engineering is Online Social Networking Sites (OSNS). Every day many people are registering with OSNS to share ideas, make friends, and do other types of activities. With the new users in the OSNS, large data is generated every day by storing users' profiles for further activities. In OSNS, security and privacy are most widely used to prevent attacks on OSNS sever and also personal profiles. It is very important to every user to have privacy for the multimedia content which is accessed by the user. Deep Learning (DL) is the most trending domain nowadays to work on OSNS. In this paper, A Hybrid Privacy for Social Media Content system is introduced to detect the type of image that is uploaded by the OSN user. The proposed system is focused on providing privacy to prevent the user's data from being attacked. The proposed system is integrated with robust pre-processing, Convolutional neural networks (CNNs), and Adaptive Privacy Policy Prediction (A3P). Results show the performance of the proposed system.