Enhancing Airport Security: Integrating PSO-TAM with Deep Learning for Real-Time Threat Assessment

To increase their threat detecting
ability in face of mounting security issues, airports are using
creative algorithms. Objective: This work merges deep learning
approaches with the Passenger Security Optimization with
Threat Assessment Model (PSO-TAM) to improve real-time
threat assessment. Contribution: This paper addresses the basic
problem of the need of a more efficient and effective security
model merging new analytics with present airport security
technologies. The challenge is especially designing a system that
reduces false alarms, increases threat detection accuracy, and
provides security workers with fast and valuable information.
Results and Findings: PSO-TAM routinely delivers superior
accuracy, precision, recall, and F-measure even if it also shows
lowered loss values. This implies that PSO-TAM not only raises
the accuracy of threat predictions but also reduces false alarms,
therefore enabling more consistent and strong security
responses. Combining PSO with the TAM has demonstrated to
be very beneficial in optimizing the classification parameters
and changing to varied data volumes.