Prediction of the IoT Security Situation Utilizing Hybrid Convolutional Neural Network and Long-Short Term Memory

The joint network of interconnected devices is known as Internet of Things (IoT), that devices are capable to decide without human intervention and all the electric devices that are collected to connect with internet in order to share the data. So the advancement of these have increased popularity tremendously over past few years. This has been achieved by interconnecting the various functionalities with the heterogeneous physical devices. They have drastically increased the human lives in order to enhance their day- to-day tasks. This increases the risk of security of the data and privacy of the users. In order to secure the data from the threats and cyber-attacks a Hybrid Convolutional Neural Network and Long-Short Term Memory (Hybrid CNN-LSTM) is proposed in order to predict the IoT security situation. For preprocessing these methods are employed elimination of the unimportant features, random selection, and the duplication removal. Then, the feature selection of done by utilizing the Decision Tree (DT) and Recursive Feature Elimination (RFE). Finally, the proposed Hybrid CNN-LSTM is utilized as a classifier. The results provide the how the efficiency of proposed Hybrid CNN-LSTM in terms of accuracy, precision, recall, and f1-score, when the existing methods are compared with the proposed method.