AcuGIS
/
AR1
1
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
AR1/docs/analysis-tools/hotspot.md

2.1 KiB
Raw Blame History

Hot Spot Analysis

Identify statistically significant clusters of high and low values using Getis-Ord Gi* statistics.

Overview

Hot spot analysis uses the Getis-Ord Gi* statistic to identify statistically significant spatial clusters. Features are classified as:

  • 99% Hot Spot: Very high values, 99% confidence
  • 95% Hot Spot: High values, 95% confidence
  • 90% Hot Spot: High values, 90% confidence
  • Not Significant: No significant clustering
  • 90% Cold Spot: Low values, 90% confidence
  • 95% Cold Spot: Low values, 95% confidence
  • 99% Cold Spot: Very low values, 99% confidence

Inputs

  • Dataset: Point or polygon dataset
  • Value Field: Numeric field to analyze
  • Neighbor Type: Distance-based or K-nearest neighbors
  • Distance (if distance-based): Maximum neighbor distance
  • K Neighbors (if KNN): Number of nearest neighbors

Outputs

New dataset containing:

  • Original geometry
  • Gi Z-Score*: Standardized z-score
  • P-Value: Statistical significance
  • Hot Spot Class: Categorized class
  • Original attributes

Algorithm

  1. Calculate spatial weights matrix based on neighbor configuration
  2. Compute Getis-Ord Gi* statistic for each feature
  3. Calculate z-scores and p-values
  4. Categorize into hot spot classes
  5. Store results in output dataset

Example

{
  "dataset_id": 123,
  "value_field": "population",
  "neighbor_type": "distance",
  "distance": 1000
}

Background Jobs

This analysis runs as a background job. See Hot Spot Analysis Worker for details.

Use Cases

  • Crime analysis
  • Disease clustering
  • Economic activity patterns
  • Environmental monitoring
  • Social phenomena analysis

Notes

  • Requires numeric field with sufficient variation
  • Distance should be appropriate for data scale
  • KNN method is generally faster for large datasets
  • Results depend on neighbor configuration

Related Documentation