The major threat to online social network (OSN) users is account compromisation. Spammers now spread malicious messages by exploiting the trust relationship established between account owners and their friends. The challenge in detecting a compromised account by service providers is validating the trusted relationship established between the account owners, their friends, and the spammers. Another challenge is the increase in required human interaction with feature selection. Research available on supervised learning has limitations with feature selection and accounts that cannot be profiled, like application programming interface (API). Therefore, this article discusses the various behaviors of OSN users and the current approaches in detecting a compromised OSN account, emphasizing its limitations and challenges. We propose a deep learning approach that addresses and resolve the constraints faced by previous schemes. We detailed our proposed optimized nonsymmetric deep autoencoder (OPT_NSDAE) for unsupervised feature learning, which reduces the required human interaction levels in the selection and extraction of features. We evaluated our proposed classifier using three different social network datasets, Facebook, Google+, and Twitter, in addition to the NSL-KDD and KDDCUP’99 datasets, in a graphical-user-interface-enabled Weka application. The experimental results show that our proposed approach outperformed most of the traditional schemes in OSN compromised account detection.

In mobile ad hoc networks (MANETs), mobile devices connect with other devices wirelessly, where there is no central administration. They are prone to different types of attacks such as the black hole, insider, gray hole, wormhole, faulty node, and packet drop, which considerably interrupt to perform secure communication. This paper has implemented the denial‐of‐service attacks like black hole attacks on general‐purpose ad hoc on‐demand distance vector (AODV) protocol. It uses three approaches: normal AODV, black hole AODV (BH_AODV), and detected black hole AODV (D_BH_AODV), wherein we observe that black holes acutely degrade the performance of networks. We have detected the black hole attacks within the networks using two techniques: (1) intrusion detection system (IDS) and (2) encryption technique (digital signature) with the concept of prevention. Moreover, normal AODV, BH_AODV, and D_BH_AODV protocols are investigated for various quality of service (QoS) parameters, i.e., packet delivery ratio (PDR), delay, and overhead with varying the number of nodes, packet sizes, and simulation times. The NS2 software has been used as a simulation tool to simulate existing network topologies, but it does not contain any mechanism to simulate malicious protocols by itself; therefore, we have developed and implemented a D_BH_AODV routing protocol. The outcomes show that the proposed D_BH_AODV approach for the PDR value delivers around 40 to 50% for varying nodes and packets. In contrast, the delay decreases from 300 to 100 ms and 150 to 50 ms with an increase in the number of nodes and packets, respectively. Furthermore, the overhead changes from 1 to 3 for various nodes and packet values. The outcome of this research proves that the black hole attack degrades the overall performance of the network, while the D_BH_AODV enhances the QoS performance since it detects the black hole nodes and avoids them to establish the communication between nodes.