How Does Graph Data Reveal Hidden Crime Networks?

Published in AGEDB , 3 min read, Jun 1

Criminal network analysis with the perspective of graph data and networks science.

Integrated information on criminals and criminal cases is essential for domestic and international criminal investigation and analysis.

However, most of this information is divided by individual and case, along with personal information protection, so that only fragmented and fragmented information can be provided in necessary situations.

Nevertheless, if graph data is used, analyzing crime-related networks that integrate heterogeneous information data is possible. In other words, it can help the efficient and effective criminal investigation by accurately recognizing the connection and importance of people or events through the graph data method.

Depending on the graph data's properties, the criminal network's nodes and links are largel classified into two types.

  • Crime networks: Based on the similarity between event data, there is an event network that defines a node as an event and a relationship between the events as an edge.

  • Criminal networks: A person network that extracts relationship information between people from person data and defines it as a person (node) and a line connecting the edges (relationships between people).

We form graphs performing duplex higher-order interactions incorporating the two network taxonomies.


After configuring data in different formats into a network, the networks are integrated to build integrated graph data through one integrated network data.

Information that can affect the inference of suspects or people who are difficult to analyze can be reflected in single data by using all information between people or information linked between events from heterogeneous graph data configured in this way.

For example :

  • Other than the suspects, who else participated in the crime?

  • Who plays what role in the criminal network?

  • Who is the most central figure in a criminal network?

By performing tasks such as, you can identify hidden correlations between each crime and the offender.

In addition, the existing network analysis methodology can be borrowed and used directly for graph data analysis to provide optimized services.

    Suppose that :

  • You can view the entire criminal network.

  • Who is the key person?

  • Who played a role in the crime?

  • How to Disrupt the Criminal Network?

etc. can be identified in a short time.


Criminal network-based crime and criminal detection using graph data analysis is a very useful technology for crime prevention and investigation, and by using it, it can increase the efficiency of crime prevention and investigation and greatly contribute to crime prevention and public safety.

In addition, the implications of helping criminal investigations by identifying related suspects based on connections between cases or identifying relationships between cases based on connected persons are the achievements achieved through the development of graph data technology.


  • E Ferrara et al., Detecting criminal organizations in mobile phone networks., Expert Systems with Applications 2014.

Fraud Detection
Use Cases
Database Technologies

The information you provide will be used in accordance with our Privacy Policy .