Ransomware is a significant cybersecurity risk that keeps increasing and threats like the WannaCry keep showing the need to be more intelligent and adaptive in detection methods. This paper presents DeepRansom, a powerful deep learning model that has been designed to detect WannaCry ransomware in complex malware environments with high precision and in a transparent manner. The architecture combines the feature selection of TabNet and the learning context of the FT-Transformer to create a dual-model architecture that can provide performance and interpretability. DeepRansom was trained and tested on a broad dataset of 21,752 samples of 26 malware families with 11 key ransomware variants. The proposed model reformulates WannaCry recognition as a binary classification task and delivers better results than the classical algorithms including Random Forest, SVM, and XGBoost among other deep learning baselines. The experiments have proven accuracy of 99.82%, precision of 97.89%, recall of 93.94% and AUC of 0.99% with strong generalization and low false positives. In addition, the SHAP-based analysis of interpretability shows that the predictions obtained by DeepRansom are based on the appropriate behavioral and structural predictors, which supports the idea of its implementation in the practice of network forensics and threat intelligence missions.
Advancements in smart surveillance and digital forensics have been significantly influenced by artificial intelligence (AI), machine learning (ML), and cloud computing, enabling real-time forensic analysis of CCTV footage to detect suspicious activities and reduce reliance on manual monitoring. While these technologies enhance public safety and policing efficiency, they also raise important concerns related to privacy, ethics, and regulation. This study proposes a deep learning-based forensic framework for real-time detection of suspicious human activity in CCTV videos, trained without relying on any external sensors. The model utilizes EfficientNet-B5 for spatial feature extraction and a Vision Transformer (ViT) that has been adapted to model both spatial and temporal dependencies between consecutive video frames. Specifically, the ViT processes frame sequences as token embeddings, enabling temporal context learning through self-attention. The EfficientNet-B5, a convolutional neural network (CNN), functions primarily as a high-level feature extractor, capturing geometric and spatial characteristics such as scale variations, aspect ratios, and motion-relevant visual patterns from individual frames. The Binary UCF-Crime dataset is used for training and validation, with class rebalancing achieved through data augmentation and weighted loss adjustment to mitigate class imbalance. Performance is evaluated using accuracy, precision, F1-score, and AUC metrics, each averaged across 5-fold cross-validation. The reported accuracy is obtained through this cross-validation protocol and benchmarked against baseline models, including standalone EfficientNet-B5 and traditional machine learning approaches. The ViT-enhanced model achieves an accuracy of 87.23%, outperforming the EfficientNet-B5 baseline by a significant margin. To address ethical concerns, the framework incorporates anonymization of personal identities and local edge-based processing to prevent raw data exposure.
Insects associated with vertebrate carrions play an important role in forensic investigations, especially in estimating post-mortem interval (PMI). Unfortunately, baseline data on local insect species associated with carrions in southern Nigeria is still limited. In this study, a checklist was compiled of insects associated with asphyxiated pig carrion was carried out in a grassland habitat. The study was conducted across both dry and rainy seasons of 2023, 2024, and 2025. Fresh pig carrions of 21kg were humanely euthanized by asphyxiation and exposed in wire-mesh cages to exclude vertebrate scavengers but allowed insects access. Insect sampling on the carrions was carried out throughout the decomposition stages. Adult and immature insects were collected, preserved, and identified to species using standard morphological keys. Four stages of carrion decomposition were observed for both seasons, but tissue depletion was faster during the rainy season than the dry season. Insect species collected on the carrions were 25 in 19 families, and 6 orders. The dipterans were the earliest and most dominant colonizers, starting from the fresh to wet stages. The coleopterans became prominent during dry stages throughout the skeletonized remains. Seasonal and interannual variations in species composition were also observed across the three-year study period. The findings provide a comprehensive checklist of carrion-associated insects in a grassland habitat in southern Nigeria and contribute baseline data for forensic entomology in the region. This study emphasizes the importance of localized entomological data for improving the accuracy of PMI estimation and forensic investigations in tropical environments.
Forensic document examination is a highly specialized branch of forensic science that focuses on determining the authenticity of disputed documents. Practitioners use a variety of scientific techniques to detect forgery, alterations, and other forms of document tampering, including ultraviolet (UV) and infrared (IR) imaging, electrostatic detection (ESDA), and high-resolution microscopy. Despite its evidentiary value, this discipline has received limited recognition in Pakistan's legal system. The study aimed to assess the role of forensic questioned documents in criminal investigation and its evidentiary limitations in the justice system of Khyber Pakhtunkhwa, Pakistan. The researcher conducted qualitative research by interviewing five forensic document examiners who work at the Peshawar Forensic Science Laboratory; along with fifteen public prosecutors, and five defense counsels. Since there are a limited number of forensic document examiners in the provincial laboratory, a census sample was used, while purposive sampling was employed for public prosecutors and defense counsels who had been involved in experiences with documents related cases. Thematic analysis of interview transcripts identified key themes, including chain of custody, process of forensic analysis, judicial perception, limited weightage of forensic reports, forensic capacity gaps. Findings revealed that the most frequent disputed documents received by the lab, were stamp papers, cheques, academic degrees, Nikah Namas (marriage certificates), burnt documents, mutation records, and even writings on walls, leaves, and human skin. Tasks such as ink differentiation, indented writing detection, and signature verification are currently performed using VSC 6000/HS and ESDA-2 apparatus however, they require more advanced instruments such as VSC 9000/HS and high-resolution microscopes for optimal results. Courts treat forensic reports as corroborative rather than primary evidence. Major challenges include delays in reports, poor stakeholder coordination, and limited qualifications and training of experts. The findings suggested that the role of forensic analysts need to be enhanced by providing them advance training. Further, policy reforms and mechanism of collaborative relationships between the relevant stakeholders for the pursuance of criminal investigation need to be developed for effective justice system administration
Incestuous paternity investigations are considered among the most analytically and ethically complex challenges in forensic genetics. Extensive allele sharing occurs among family members and close biological relatives. Mutations in short tandem repeat (STR) markers, together with the concealment or frequent absence of key family members, can interfere with reliable paternity interpretation and increase the risk of false inclusion or exclusion. This study describes a legally significant forensic DNA case, where the use of the national DNA database revealed unexpected paternal candidates, ultimately resolving a highly complex case. Initial autosomal STR profiling using AmpFlSTR™ Identifiler™ Plus and GlobalFiler™ multiplex systems indicated non-exclusion of an alleged father (AF1) accused of sibling incest, despite a single-locus mismatch consistent with mutation. Subsequent blind comparison within the DNA Database identified a second related individual who also could not be excluded as the biological father (AF2). Further investigation, including Y-chromosomal STR analysis and expanded testing of multiple maternal relatives, ultimately established a third individual (AF3), a maternal nephew, as the true biological father without mutational inconsistencies. These findings highlight the essential role of mutation-aware interpretation, comprehensive kinship analysis, and forensic DNA databases in accurately resolving complex incest-associated paternity disputes and preventing wrongful judicial outcomes.
