Harmonising Heterogeneous Data Sources for Comprehensive Forensic Analysis
Pradeep Kumar Chandra, Modi Himabindu, Vijilius Helena Raj, Amit Dutt, Koreswar Rao Kumbha, Abdulkareem Mahdy · 2024
New techniques must be devised to integrate data types for extensive and accurate analysis as digital forensics evolves. Five key algorithms are recommended for a full framework in this study. Cross-platform normalization for standardization, semantic integration for common representation, federated learning for collaborative analysis, ontology-driven anomaly detection, and blockchain-enhanced chain of custody. The semantic integration algorithm creates a common vocabulary, maps data sources, and standardizes data first. Future algorithms employ this common representation to handle standardization, joint analysis, anomaly detection, and ownership tracking concerns. The ablation research details what each software undertakes to improve and adapt to real-time changes. The proposed performance evaluation system is more accurate, safer, customizable, and effective than present techniques. Federated learning allows numerous analysts to work together while protecting data, while ontology-driven anomaly detection detects anomalies with data consistency. Blockchain-Enhanced Chain of Custody makes Common Representation more dependable by employing a safe and independent blockchain. Forensic research in evolving data systems presents challenges, which this paradigm addresses. The offered methodologies give specialists additional tools to combine and assess multiple types of data, providing a full and flexible solution for current forensic investigations.