Enhancing Threat Detection Using Artificial Intelligence in Modern Cybersecurity Systems Using SPSS Statistics
Rajendar Dommeti
Abstract
The proactive identification of threats, response to incidents, and prediction of patterns of future cyber threats by organizations using machine learning and deep learning techniques have made artificial intelligence a game-changing technology. The limitations of traditional security systems, which face challenges in detecting complex attacks in real time, have been demonstrated by the rapid pace of evolution of cyber threats. The importance of this work lies in the fact that it addresses the emerging gap between traditional security techniques and the current cyber threats. By systematically reviewing AI's role in threat detection, behavioral analytics, and vulnerability prediction, it provides actionable insights for IT professionals, security analysts, and organizations seeking intelligent, adaptive, and next-generation cybersecurity frameworks. The study collected data from 100 respondents across diverse industries and job roles, evaluating eight AI security tools including Darktrace, CrowdStrike, and IBM QRadar. Five key parameters — Detection Accuracy, Response Speed, False Positive Rate, System Integration, and Effectiveness — were assessed. IBM SPSS Statistics was used for reliability testing, descriptive analysis, frequency distribution, and correlation analysis. Cronbach's Alpha of 0.752 confirmed scale reliability. Mean scores ranged from 3.58 to 3.74, reflecting moderately high perceptions across all variables. Correlation analysis revealed that System Integration and Effectiveness shared the strongest relationship (r = 0.512), with all variables showing statistically significant positive correlations.