Publication Search

Suspicious urban sound recognition is a critical component of intelligent public safety and urban monitoring systems, enabling the automated identification of anomalous acoustic events such as gunshots, sirens, and other security-sensitive sounds. However, existing deep learning approaches often struggle to simultaneously capture long-range temporal dependencies and global contextual relationships, particularly under noisy and acoustically complex urban conditions. This limitation can reduce reliability in safety-critical scenarios where missed detections carry significant risk. To address these challenges, this study proposes a Multi-Branch Bidirectional Long Short-Term Memory (BiLSTM) framework with Multi-Head Self-Attention (MHSA) for enhanced sequential and contextual feature modeling. Mel-frequency cepstral coefficients (MFCCs) are extracted from a curated subset of the UrbanSound8K dataset, comprising five suspicious sound classes, and used as input to the proposed architecture. The multi-branch design enables complementary temporal representations, while the self-attention mechanism provides lightweight contextual weighting of BiLSTM outputs. Experimental results demonstrate that the proposed model achieves a test accuracy of 95.59%, outperforming conventional Dense and LSTM-based baseline models under identical experimental settings. An ablation study further confirms the contribution of multi-branch integration and attention-based enhancement to overall performance. Class-wise evaluation reveals consistently high recall across all sound categories, particularly for safety-critical classes such as gunshots and sirens. These findings indicate that the proposed framework provides robust and reliable performance, making it suitable for real-time smart city surveillance and public safety applications.

Honor killing cases often involve a distributed structure of perpetrators between decision-makers, providers of means, and implementers. This collective pattern raises the problem of role attribution in criminal law enforcement, which often shifts toward two problematic tendencies: centralizing responsibility on the executor or expanding criminal responsibility based on family ties. This article aims to formulate a tested role attribution model so that criminal responsibility does not stop at the direct perpetrator and does not develop into association-based punishment. This research uses a normative legal research method with a conceptual approach. Data collection methods were collected using literature studies, then analyzed qualitatively and presented descriptively. The research results formulate a role map of instigator, facilitator, and executor, operationalized through group role attribution based on two axes: causal contribution and normative contribution. The instigator is understood as the driver who shapes the will and locks the decision, the facilitator is understood as an assistant who deliberately provides the opportunity, means, or information. Meanwhile, the executor is someone who carries out the material act, although in terms of position, their actions are not automatically identical to the dominance of the decision. This division of roles is complemented by evidentiary indicators covering communication, financing, provision of facilities, field control, and post-incident intimidation, along with negative criteria to prevent inferences based on blood relations or passive presence. This model provides a more measurable standard of attribution for investigation, prosecution, and sentencing in collective honor killing cases.