Nearest neighbour analysis

03-09-2021 10:05 AM
New Contributor


I am trying to use (nearest neighbor analysis) in the context of malaria cases, and I have the following inquiries please to help me understand it as am building my knowledge about it and the online information I came across was confusing:

  • Can I have good thorough and easy to understand resources explaining: Average nearest neighbor (ANN) analysis, and other spatial analyses used as an alternative for spatial clustering identification and quantification, please?
  • what is the difference between ANN, K-nearest neighbor (KNN), and K-function?
  • is "K function" is the same as "KNN"??Here is an article about KNN and a link explaining K-function, I am not sure if they are the same thing or different?
  • what is the difference between ANN and KNN? 
  • KNN do assess cluster from a time and space perspective, meaning it assess clustering by evaluating whether the points/cases are near in location and whether the points are nearby by the time of occurrence.
    • wouldn't that require my data to have the exact time each infection case occurred? 
    • Does ANN consider both "Time" and space? and if yes, would having all cases diagnosed at the SAME time through a cross-sectional survey be a limitation to the ANN test?  
  • what are the assumptions underlying and controlling ANN? and its limitations?
I appreciate any feedback and thank you for your time and support.
0 Kudos
1 Reply
MVP Esteemed Contributor

Have you looked at the offerings with the help documents under the Spatial Statistics toolsets?

An overview of the Spatial Statistics toolbox—ArcGIS Pro | Documentation

A couple of sample links

An overview of the Analyzing Patterns toolset—ArcGIS Pro | Documentation

How Average Nearest Neighbor works—ArcGIS Pro | Documentation

... sort of retired...
0 Kudos