Intrusion Detection Attacks Classification using Machine Learning Techniques

Abstract

Distributing numerous services over the internet is called Cloud Computing. Applications and tools like networking, data storage, databases, servers, software are examples of the resources. The service provider is required to provide the resource always and from any location. However, the network is the most important factor in gaining access to data in the cloud. When leveraging the cloud network, the cloud threats take advantage. An intrusion Detection System (IDS) observes the network and detects and reports threats. The anomaly method is significant in Intrusion Detection Systems. IDS monitors known and unknown data whenever a virtual machine is developed. If any anonymous data is detected, the Intrusion Detection System identifies it using an anomaly classification algorithm and sends a report to the administrator. Naive Bayes, Decision tree (CART), Support Vector Machine, and random forest techniques are utilized in this work to classify unknown data. These algorithms are assisting in reducing the percentage of false alarms. This proposed work was carried out utilizing the WEKA tool for generating the report, yielding a best result in less computing time.