# getHotspots: find spatial clusters using supervised learning methods In TreeHotspots: Hotspot Detection using Classification Trees

## Description

Here we consider one fixed rotation of the data and

seek hot spots using a classification tree

## Usage

 ``` 1 2 3 4 5 6 7 8 9 10``` ```getHotspots(formula = violent ~ X + Y, rotX, NullClass = "0", minsize = 200, minArea = 20, maxArea = 250, ORfilter = list(OR = TRUE, OR1 = 1.8, OR2 = 0.1), TreeAlgorithm = c("rpart", "ctree")[1], verbose = 1, ...) ```

## Arguments

 `formula` A formula expression. The left-hand-side (response) should be either a numerical vector when a regression tree will be fitted or a factor, when a classification tree is produced. The right-hand-side should be a series of numeric or factor variables separated by +; there should be no interaction terms. Both . and - are allowed: regression trees can have offset terms. `rotX` data, possibly already rotated `NullClass` if y is a factor, this is the category used for the background `minsize` minimum number of points inside a cluster `minArea` minimum area of a cluster, units: lat*lon `maxArea` maximum area of a cluster, units: lat*lon `ORfilter` filter on minimum and maximum odds ratios (OR) `TreeAlgorithm` which tree algorithm to choose `verbose` level of verbosity `...` further arguments to `tree`

## Value

identified clusters (if any)

Markus Loecher

## Examples

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28``` ```#examples to come data("drugCrimes", envir = environment()) drugCrimes\$MATCH = factor(drugCrimes\$MATCH) spot1 = getHotspots(MATCH ~ X+Y,drugCrimes, minArea=20/10^6,maxArea=250/10^6, ORfilter=list(OR=FALSE,OR1=0.8,OR2=0.1)) suppressWarnings(suppressMessages(library("PBSmapping"))) PBSmapping::plotPolys(spot1[1:5,],density=NULL,xlim=range(drugCrimes\$X),ylim=range(drugCrimes\$Y), border="blue",lwd=2) ranRows=sample(1:nrow(drugCrimes), 5000) points(Y~X,data=drugCrimes[ranRows,],col=RgoogleMaps::AddAlpha(4-as.numeric(MATCH)),pch=20,cex=0.6) ```

TreeHotspots documentation built on May 31, 2017, 3:01 a.m.