Security-Aware Information Classification Using Attributes Extraction for Big Data Cyber Security Analytics

Fast development of web-based solutions with different paradigms and multiple industries having platforms emerges as enablers or value creators. Financial companies also have enabled by trends, help develop new services, and enhance internal business procedures, to enhance the now-optional data sharing of realized value enhancements between financial institutions. At the same time, there were also concerns about the data leakage regarding information protection which affects both customers and financial organizations. In order to identify the proper data classifications, it is critical for stakeholders in the financial services industry to determine what information can be shared across financial services companies. In order to detect cyber-attacks that big data cybersecurity analytics (BDCA) systems impact those big data technologies (for example, Spark and Hadoop) for store, analyze, and collect large amounts of security incident data, this research focuses on this problem by describing a new approach that combines attributes extraction for large data cybersecurity analytics. For BDCA systems, reaction time and accuracy are the two most important factors. Frequent modifications to the BDCA system's operating environment, such as those affecting the quantity and quality of security data events, have a substantial impact. Lastly, the technique is evaluated in single-mode and multimode settings utilizing various adaption scenarios and a Hadoop-based BDCA system. The response time and accuracy of the average BDCA are enhanced, indicating evaluation.