Enhancing Binary Encoded Crime Linkage Analysis Using Siamese Network

2025-11-10

AI Summary

To address low accuracy in serial crime linkage caused by high-dimensional, sparse, and heterogeneous crime data, this paper proposes a Siamese autoencoder model integrated with geographic-temporal features. Innovatively, geographic-temporal information is incorporated into the decoder stage—rather than the input layer—to prevent dilution of behavioral signals by sparse raw features. A domain-knowledge-guided dimensionality reduction strategy further enhances representation interpretability and practical utility. The model employs contrastive learning to capture behavioral similarity between crime instances and adopts binary encoding to strengthen discriminative capability. Experiments on the real-world forensic dataset ViCLAS demonstrate a 9% improvement in AUC over conventional methods, significantly boosting serial crime association accuracy. This work provides a robust technical foundation for intelligent crime linkage analysis and investigative decision support.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Deep Generative Models & Autoencoders
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Effective crime linkage analysis is crucial for identifying serial offenders and enhancing public safety. To address limitations of traditional crime linkage methods in handling high-dimensional, sparse, and heterogeneous data, we propose a Siamese Autoencoder framework that learns meaningful latent representations and uncovers correlations in complex crime data. Using data from the Violent Crime Linkage Analysis System (ViCLAS), maintained by the Serious Crime Analysis Section of the UK's National Crime Agency, our approach mitigates signal dilution in sparse feature spaces by integrating geographic-temporal features at the decoder stage. This design amplifies behavioral representations rather than allowing them to be overshadowed at the input level, yielding consistent improvements across multiple evaluation metrics. We further analyze how different domain-informed data reduction strategies influence model performance, providing practical guidance for preprocessing in crime linkage contexts. Our results show that advanced machine learning approaches can substantially enhance linkage accuracy, improving AUC by up to 9% over traditional methods while offering interpretable insights to support investigative decision-making.

Problem

• Enhancing crime linkage accuracy using Siamese networks for serial offender identification
• Addressing high-dimensional sparse data limitations in traditional crime analysis methods
• Improving behavioral pattern recognition in crime data through geographic-temporal integration

Innovation

• Siamese Autoencoder framework for crime linkage analysis
• Geographic-temporal feature integration at decoder stage
• Domain-informed data reduction strategies for preprocessing guidance