A Novel Hybrıd Approach for Cyber Securıty ın Iot Inetwork usıng Deep Learnıng Technıques
In this paper we propose deep learning models for the cyber security in IoT (Internet of Things) networks. IoT network is as a promising technology which connects the living and nonliving things around the world. The implementation of IoT is growing fast but the cyber security is still a loophole, so it is susceptible to many cyber-attacks and for the success of any network it most important that the network is completely secure, otherwise people could be reluctant to use this technology. Therefore, a dire need emerges to find a method to extract meaningful information from big data, classify it into different categories and predict end user’s behaviors or sentiments. Long Short-Term Memory (LSTM) model and Convolutional Neural Network (CNN) model have been applied to different Natural Language Processing (NLP) tasks with remarkable and effective results. The CNN model efficiently extracts higher level features using convolutional layers and max-pooling layers. The LSTM model is capable to capture long-term dependencies between word sequences. In this study, we propose a hybrid model using LSTM and very deep CNN model named as Hybrid CNN-LSTM Model to overcome the sentiment analysis problem. Our results show that the proposed Hybrid CNN-LSTM Model outperforms traditional deep learning and machine learning techniques in terms of precision, recall, f-measure, and accuracy. DDoS (Distributed Denial of Service) attack has affected many IoT networks in recent past that has resulted in huge losses. We have proposed deep learning models and evaluated those using latest CICIDS2017 datasets for DDoS attack detection which has provided highest accuracy as 97.16% also proposed models are compared with machine learning algorithms. Convolutional Neural Networks (CNNs) have become the defector standard for several Computer Apparition and Machine Learning operations. the modest and compacted configuration of convolutions scalar multiplications and additions. This paper also identifies open research challenges for usage of deep learning algorithm for IoT cyber security.