Intelligent Network Forensics: AI-Powered Cybercrime Investigation using African Vultures Optimization

As the complexity of cyber threats continues to increase, intelligent network forensics is the only option available to conduct precise cybercrime investigation. In this paper, an AI-based forensic framework using Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) networks, and the African Vultures Optimization Algorithm (AVOA) is suggested to improve threat classification, feature selection, and forensic intelligence. The CNN-LSTM model is tasked with extracting spatial-temporal patterns in traffic networks, whereas AVOA is utilized for adaptive hyperparameter tuning and best feature selection. The designed system overcomes traditional challenges in forensic analysis, including high false positives and lack of adaptability to emerging threats. Experimental assessment on the UNSW-NB15 dataset achieves a 98.9% classification rate, a far lower false positive rate, and better detection efficiency than existing models. This study provides a scalable, adaptive, precision-driven cybercrime detection solution, delivering actionable intelligence via AI-tuned digital forensics.