A Predictive Analytics Approach to Traceability in Digital Forensics Using Time Series Analysis

Digital forensics is essential to cybersecurity and the examination of virtual activities –particularly criminal ones. This research outlines a predictive analytical technique in his based on the time series to improve digital forensics traceability. The performance of the research is assessed by using the Digital Forensic Investigation Dataset (DFID) which contains application logs, network traffic files and file system images derived from inv You can download simulated forensic cases which are time stamped application logs, network traffic records and file system snapshots which are used to evaluate the performance of the research through Digital Forensic Investigation Dataset (DFID). In this regard, machine learning and deep learning algorithms such as Random Forest, Support Vector Machines, Long Short-Term Memory Networks were used to anticipate and detect the notion of suspicious action. Concerning the results of the Random Forest model, it yielded a high accuracy of 95% and precision of 94%, along with a recall of 96%, F1-score of 95%, and AUC of 0. 98. The feature importance analysis emphasized the most important features in the forensic process such as time, Internet protocol addresses, file hashes, and types of events. This research contributes to enhancing the knowledge base in the field of digital forensics by proposing the use of predictive analytics for early identification and mitigation of threats; this forms part of making Investigations efficient. The results support the notion that informational knowledge, as well as currently practiced and emerging computational methods, should play a central role in forensic practices. Some of the areas of further work are the attempt to combine the approaches and the refinement of the deep learning schemes to faces more complicated circumstances that can be encountered in forensics, as well as the integration of legal and ethical issues to the use of predictive models within digital environment of forensic investigations.