COMPARATIVE ANALYSIS OF MACHINE LEARNING MODELS FOR DETECTING EVASIVE SMS SPAM
Keywords:
Genetic Algorithm (GA), Convolutional Neural Network (CNN), Hyperparameter Optimization, Driver Drowsiness Detection, Real-Time Monitoring.Abstract
Convolutional neural networks are enhanced by genetic algorithms to detect drowsy drivers. Detection enhances model performance by locating the optimal hyperparameters. In this approach, Convolutional Neural Networks (CNNs) and Genetic Algorithms (GAs) are employed to modify learning rates, filter dimension, and layer layout. By repeatedly modifying CNN parameters through mutation, selection, and crossover, genetic algorithms enhance detection performance. Eye closing and yawning patterns are crucial facial signals that the upgraded CNN records for real-time fatigue measurement. In addition to reducing handling costs and overfitting, this strategy makes items more durable. The results demonstrate that, compared to conventional CNNs, the convergence process is quicker and more accurate. This system improves traffic safety by consistently monitoring drivers. Researchers will primarily focus on real-world applications and ways to integrate many sensors in the future.
