Research on the application of network security defence in database security services based on deep learning integrated with big data analytics

Abstract

Every day, more people use the internet to send and receive sensitive information. A lot of confidential data is being transmitted electronically between people and businesses. Cyber-attacks, which are the inevitable result of our growing reliance on digital technology, are a reality that we must face today. This paper aims to investigate the impact of Big Data Analytics (BDA) on information security and vice versa. Additionally, an Artificial Neural Network (ANN)-based Deep Learning (DL) method for Anomaly Detection (AD) is presented in this work. To improve AD, the proposed method uses a DL-based detection method, which is used to parse through many collected security events to develop individual event profiles. The paper also investigated how BDA can be used to address Information Security (IS) issues and how existing Big Data technologies can be adapted to improve BDA's security. This study developed a DL-based Security Information System (DL-SIS) using a combination of event identification for data preprocessing and different Artificial Neural Network (ANN) methods. The feasibility and impact of implementing a Big Data Analytics (BDA) system for AD are investigated and addressed in this study. From this study, we learn that BDA systems are highly effective in securing Critical Information Setup from several discrete cyberattacks and that they are currently the best method available. By analyzing the False Positive Rate (FPR), the system facilitates quick action by security analysts in response to cyber threats. DL-SIS had the highest AD accuracy of 99.40% but performed poorly in the high-dimensional dataset.