Crime Analysis

Repeat and Near Repeat Classification

Summary

Use a series of distance and time values to classify incidents as originators, repeats, or near-repeats, and to identify potential spatial and temporal relationships between incidents.

Usage

  • The Input Features must be a point feature class or shape file with a date field, representing the locations of incidents.
  • The tool will use all points with date values. Incidents without valid dates will be excluded and a warning message will print to the screen listing the skipped incidents.
  • This tool will honor a selected set of features. When the Input Features contain a selection, only the selected features will be considered when classifying features and building connections.
  • Incidents are classified as O (Originator), NR (near-repeat), or R (repeat) according to their proximity to each other in space and time using each combination of the specified spatial and temporal band values, and the specified repeat distance.
    • Incidents are classified as originators if they are the originator of at least one other incident and have no preceding incidents within the current spatial and temporal ranges.
    • Incidents are classified as near-repeat incidents when they follow a previous repeat, near-repeat, or originating incident within the current spatial and temporal ranges, and do not meet the criteria to be classified as a repeat incident.
    • Incidents are classified as repeat incidents when they occur within the specified repeat distance and the current temporal range. An incident that meets the requirements for both a repeat and a near-repeat incident is classified as a repeat.
    • Incidents that do not meet the criteria for any of the above are left unclassified.
  • This tool creates three outputs:
    • Create a copy of the input incident features with classification fields added. These fields are populated with the classification values for each incident within each set to spatial and temporal ranges.

      For example, if a repeat distance of 1 map unit (Feet), spatial bands of 100 and 200 map units (Feet), and temporal bands of 7 and 14 days are specified, the tool will append 6 fields with the following field names and aliases:

      • sb0tb1 (1 ft / 7 Days): Classification of incidents that are originators of repeat incidents occurring no more than 7 days later, and incidents that occurred within the repeat distance and 7 days of a previous incident.
      • sb0tb1 (1 ft / 14 Days): Classification of incidents that are originators of repeat incidents occurring no more than 14 days later, and incidents that occurred within the repeat distance and 14 days of a previous incident.
      • sb1tb1 (100 ft / 7 Days): Classification of incidents that are originators of repeat and near-repeat incidents occurring no more than 7 days later, and incidents that occurred within 100 map units and 7 days of a previous incident.
      • sb1tb2 (100 ft / 14 Days): Classification of incidents that are originators of repeat and near-repeat incidents occurring no more than 14 days later, and incidents that occurred within 100 map units and 14 days of a previous incident.
      • sb2tb1 (200 ft / 7 Days): Classification of incidents that are originators of repeat and near-repeat incidents occurring no more than 7 days later, and incidents that occurred within 200 map units and 7 days of a previous incident.
      • sb2tb2 (200 ft / 14 Days): Classification of incidents that are originators of repeat and near-repeat incidents occurring no more than 14 days later, and incidents that occurred within 200 map units and 14 days of a previous incident.
      • The tool also appends a field called ZVALUE with values representing the number of days between each incident and the smallest date value in the dataset. These values can be used to visualize the time sequence of incidents in ArcGIS Pro so that oldest incidents appear near the ground and more recent incidents appear further from the ground.
    • Creates a line feature class that is populated with z-enabled line features representing the potential relationships between repeat and near-repeat incidents and their originators. The z values for each vertex correspond to the ZVALUE field value for the incident that occurs at that vertex.
    • Creates a comma-separated values (CSV) file summary report with the number of incidents processed, the counts and proportions of incidents that fall within each spatial and temporal band combination, and the estimated half-life and half-distance values which can be used as inputs for the Calculate Prediction Zones tool.
  • All distances are calculated using geodesic measurements.

Syntax

RepeatNearRepeatClassification_crime (Input_Feature_Class, Output_Incident_Points, Output_Incident_Connection_Lines, Output_Summary_Report_Location, Date_Field, Repeat_Incident_Distance, Spatial_Bands, Temporal_Bands)

Parameter Explanation Data Type
Input Feature Class Feature class containing points representing the location of incidents to classify. The data contained in the features class will typically cover a long time period (e.g., one year). The feature class must have a date field (in date format) and all features must have date values. Feature Layer
Output Incident Points Output feature class containing calculated classification values. Feature Layer or Feature Class
Output Summary Report Location The results from the analysis of repeats and near repeats will be written into a Summary report (csv file format) for review. The report can be found in this folder after the tool has completed successfully. The results are also written to the Messages section of the Results window. Folder
Date Field The field in the input feature class containing the date on which each incident occurred. If the date of the incident spans a range (e.g., date information is recorded in two fields representing the from and to dates), choose the date field you consider to be most appropriate. This will usually be the date at the beginning of the date range. Values in the date field are used to calculate the number of days between each incident. Field
Repeat Incident Distance Maximum distance in the units of the input feature class where adjacent incidents are considered repeats rather than near repeats. A value of 0 is exact to the accuracy of the coordinate system. If the geocoding of the input feature class is likely to very slightly vary it is recommended this value is set to 1 (i.e., incidents will be classified as repeats if the distance between them is less than or equal to 1 map unit). Double
Spatial Bands Distance in the units of the input feature class to classify near incidents. Multiple bands can be entered. For example, bands of 200m and 400m will classify features that are located up to and including 200m from a previous incident, and up to and including 400m of a previous incident. Multiple Value
Temporal Bands Number of days to classify near incidents. Multiple bands can be entered. For example, bands of 7, 14 and 21 days will classify features that took place up to and including 7 days, 14 days, and 21 days of a previous incident. Multiple Value

Licensing information

  • ArcGIS Desktop Basic: No
  • ArcGIS Desktop Standard: No
  • ArcGIS Desktop Advanced: Yes
Top