Analyses of crime patterns in NIBRS data based on a novel graph theory clustering method: Virginia as a case study

Area of Science:

• Criminology
• Data Science
• Statistical Analysis

Background:

• Analyzing crime data is crucial for law enforcement and public safety.
• Existing methods may not fully capture complex crime patterns and correlations.
• The National Incident-Based Reporting System (NIBRS) offers rich data for detailed analysis.

Purpose of the Study:

• To propose a novel clustering method for NIBRS data analysis.
• To determine correlations between different crime types and parameters.
• To develop a crime likelihood index and cluster jurisdictions by crime patterns.

Main Methods:

• Development of a novel clustering algorithm tailored for NIBRS data.
• Correlation analysis of crime types and crime parameters.
• Application of the method to 2005 Virginia assault data from 121 jurisdictions.

Main Results:

• Identified significant correlations between different crime types.
• Found correlations between crime parameters and specific crime types.
• Revealed that certain Virginia jurisdictions share distinct crime patterns.
• Demonstrated the utility of the method in understanding crime occurrences.

Conclusions:

• The novel clustering method effectively analyzes NIBRS data.
• The approach aids in understanding crime type correlations and jurisdictional patterns.
• Findings can inform crime prevention strategies and resource allocation by predicting crime likelihood.