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