New Hybrid Intrusion Detection and Prevention System to Ensure Security and Privacy in Big Data

In recent years, with the concept of IoT (Internet of Things) entering our lives thanks to the rapidly increasing number of digital applications, data is collected from a wide variety of sources at an astonishing rate and the amount of data is increasing exponentially. Such as social networks, cloud computing, and data analytics, make today possible to collect huge amounts of data. Big data has emerged as a concept at this point. Big data is the form of all the data collected from different sources such as blogs, photos, videos, log files. Therefore, big data is an important topic in many fields. However, it is not only very difficult to store big data and analyse, but also bring serious threat to the security of individual's sensitive information. This paper describes the issues surrounding big data security and privacy and provides the solution with hybrid models. Our proposed model can perform the best prediction accuracy and performance criteria. This study is based on intrusion detection and prevention systems built on big data and classifies the methods used to analyze and evaluate successes comparatively. In this paper, three different types of hybrid models are developed by "Classification + Clustering" and "Clustering + Classification" and multiple hyrid model ("Classification + Classification + Clustering"). Moreover, the three hybrid models we proposed were developed using classification and clustering methods. A real world dataset from a production company in Turkey. The experimental results show that multiple hybrid model based can provide the highest prediction accuracy and maximize the profit. Hybrid models gave more accurate results in a shorter time than other methods.