Cybersecurity Threat Detection Analysis via Exploratory Data Analysis

In this research, where the landscape of cybersecurity is rapidly evolving, the necessity for efficient threat analysis and mitigation is of paramount importance. To this end, exploratory data analysis (EDA) plays a crucial role in understanding and visualizing various patterns that can enhance the ability to detect threats through firewalls. The focus is on a financial institution, utilizing data from events spanning two years (01/01/2022 - 31/12/2023), comprising a total of 2,884,103,138 events. After filtering, 53,273,203 events were selected to identify the top 20 most frequently occurring threats. These insights allow for the formulation of hypotheses in this context to help assess risk levels and detect anomalies. The model identifies outliers by comparing specific data points to others, and the distribution is examined using histograms. This approach ensures that the strategies employed are robust and adaptable to emerging threats. Together, these methodologies form a comprehensive framework for effectively safeguarding digital infrastructure against increasingly complex cyber threats.