Decision Tree Based Interference Recognition for Fog Enabled IOT Architecture

The cyber-attacks nowadays are becoming more and more erudite causing challenges in distinguishing them and confining. These attacks affect the sensitized information’s of the network by penetrating into the network and behaving normally. The paper devises a system for such interference recognition in the internet of things architecture that is aided by the FOG. The proposed system is a combination of variety of classifiers that are founded on the decision tree as well as the rule centered conceptions. The system put forth involves the JRip and the REP tree algorithm to utilize the features of the data set as input and distinguishes between the benign and the malicious traffic in the network and includes an decision forest that is improved with the penalizing attributes of the previous trees in the final stage to classify the traffic in the network utilizing the initial data set as well as the outputs of the classifiers that were engaged in the former stages. The proffered system was examined using the dataset such BOT-Internet of things and the CICIDS2017 to evince its competence in terms of rate of false alarm, detection, and accuracy. The attained results proved that the performance of the proposed system was better compared to the exiting methodologies to recognize the interference.