Social Network Based Privacy Data Optimization Using Ensemble Deep Learning Architectures

Abstract

As a result of changes in technology, enormous amounts of data are produced every second. A tremendous amount of data is generated every second by social networking and data mining. This paper proposed the novel intrusion detection based data optimization by deep learning techniques. Initially the security has been improved by using Ensemble algorithm combining decision tree with clustering based IDS. The network generates data at a high rate, volume, and variety, making it highly challenging to identify assaults using conventional methods. So the data has been optimized using KNN classification which improves the data accuracy. We developed a neural network for detecting malicious user attacks using a deep-learning technique. We discovered that a deep learning model could improve accuracy such that a social network's ability to mitigate attacks is as effective as possible.