USArrests (clusters)

Contexto

La base de datos “USArrests” contiene estadísticas en arrestos por cada 100,000 residentes por agresión, asesinato y violación en cada uno de los 50 estados de EE.UU. en 1973.

Paso 1. Instalar paquetes y llamar librerías

#install.packages("cluster")
library(cluster)
#install.packages("ggplot2")
library(ggplot2)
#install.packages("data.table")
library(data.table)
#install.packages("factoextra")
library(factoextra)

## Welcome! Want to learn more? See two factoextra-related books at https://goo.gl/ve3WBa

library(maps)

## 
## Attaching package: 'maps'

## The following object is masked from 'package:cluster':
## 
##     votes.repub

library(tibble)
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:data.table':
## 
##     between, first, last

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

Paso 2. Obtener los datos

datos <- USArrests
head(datos)

##            Murder Assault UrbanPop Rape
## Alabama      13.2     236       58 21.2
## Alaska       10.0     263       48 44.5
## Arizona       8.1     294       80 31.0
## Arkansas      8.8     190       50 19.5
## California    9.0     276       91 40.6
## Colorado      7.9     204       78 38.7

summary(datos)

##      Murder          Assault         UrbanPop          Rape      
##  Min.   : 0.800   Min.   : 45.0   Min.   :32.00   Min.   : 7.30  
##  1st Qu.: 4.075   1st Qu.:109.0   1st Qu.:54.50   1st Qu.:15.07  
##  Median : 7.250   Median :159.0   Median :66.00   Median :20.10  
##  Mean   : 7.788   Mean   :170.8   Mean   :65.54   Mean   :21.23  
##  3rd Qu.:11.250   3rd Qu.:249.0   3rd Qu.:77.75   3rd Qu.:26.18  
##  Max.   :17.400   Max.   :337.0   Max.   :91.00   Max.   :46.00

Paso 3. Escalar las variables

df <- scale(datos)

Paso 4. Optimizar la cantidad de grupos

La cantidad óptima de grupos corresponde al punto más alto de la siguiente gráfica:

set.seed(123)

optimizacion <- clusGap(df, FUN=kmeans, nstart=1, K.max = 8)
plot(optimizacion, xlab = "Número de clusters k")

Paso 5. Cantidad de grupos

grupos <- 4

Paso 6. Generar los segmentos

segmentos <- kmeans(df, grupos)
segmentos

## K-means clustering with 4 clusters of sizes 13, 8, 16, 13
## 
## Cluster means:
##       Murder    Assault   UrbanPop        Rape
## 1 -0.9615407 -1.1066010 -0.9301069 -0.96676331
## 2  1.4118898  0.8743346 -0.8145211  0.01927104
## 3 -0.4894375 -0.3826001  0.5758298 -0.26165379
## 4  0.6950701  1.0394414  0.7226370  1.27693964
## 
## Clustering vector:
##        Alabama         Alaska        Arizona       Arkansas     California 
##              2              4              4              2              4 
##       Colorado    Connecticut       Delaware        Florida        Georgia 
##              4              3              3              4              2 
##         Hawaii          Idaho       Illinois        Indiana           Iowa 
##              3              1              4              3              1 
##         Kansas       Kentucky      Louisiana          Maine       Maryland 
##              3              1              2              1              4 
##  Massachusetts       Michigan      Minnesota    Mississippi       Missouri 
##              3              4              1              2              4 
##        Montana       Nebraska         Nevada  New Hampshire     New Jersey 
##              1              1              4              1              3 
##     New Mexico       New York North Carolina   North Dakota           Ohio 
##              4              4              2              1              3 
##       Oklahoma         Oregon   Pennsylvania   Rhode Island South Carolina 
##              3              3              3              3              2 
##   South Dakota      Tennessee          Texas           Utah        Vermont 
##              1              2              4              3              1 
##       Virginia     Washington  West Virginia      Wisconsin        Wyoming 
##              3              3              1              1              3 
## 
## Within cluster sum of squares by cluster:
## [1] 11.952463  8.316061 16.212213 19.922437
##  (between_SS / total_SS =  71.2 %)
## 
## Available components:
## 
## [1] "cluster"      "centers"      "totss"        "withinss"     "tot.withinss"
## [6] "betweenss"    "size"         "iter"         "ifault"

Paso 7. Asignar el grupo al que pertenece cada observación

asignacion <- cbind(cluster = segmentos$cluster, datos)
head(asignacion)

##            cluster Murder Assault UrbanPop Rape
## Alabama          2   13.2     236       58 21.2
## Alaska           4   10.0     263       48 44.5
## Arizona          4    8.1     294       80 31.0
## Arkansas         2    8.8     190       50 19.5
## California       4    9.0     276       91 40.6
## Colorado         4    7.9     204       78 38.7

#asignacion$cluster <- as.factor(asignacion$cluster)
#summary(asignacion$cluster)

Paso 8. Graficar los clusters

fviz_cluster(segmentos, data = datos)

Paso 9. Comparar segmentos

promedio <- aggregate(asignacion, by = list(asignacion$cluster), FUN=mean)
promedio

##   Group.1 cluster   Murder   Assault UrbanPop     Rape
## 1       1       1  3.60000  78.53846 52.07692 12.17692
## 2       2       2 13.93750 243.62500 53.75000 21.41250
## 3       3       3  5.65625 138.87500 73.87500 18.78125
## 4       4       4 10.81538 257.38462 76.00000 33.19231

Paso 10. Colorear el mapa

USArrests <- rownames_to_column(asignacion, "State")
head(USArrests)

##        State cluster Murder Assault UrbanPop Rape
## 1    Alabama       2   13.2     236       58 21.2
## 2     Alaska       4   10.0     263       48 44.5
## 3    Arizona       4    8.1     294       80 31.0
## 4   Arkansas       2    8.8     190       50 19.5
## 5 California       4    9.0     276       91 40.6
## 6   Colorado       4    7.9     204       78 38.7

cluster1 <- filter(USArrests, cluster == 1)
cluster2 <- filter(USArrests, cluster == 2)
cluster3 <- filter(USArrests, cluster == 3)
cluster4 <- filter(USArrests, cluster == 4)

map(database = "state")
map(database = "state", regions = cluster1$State, col="green", fill=TRUE, add = TRUE)
map(database = "state", regions = cluster2$State, col="red", fill=TRUE, add = TRUE)
map(database = "state", regions = cluster3$State, col="yellow", fill=TRUE, add = TRUE)
map(database = "state", regions = cluster4$State, col="orange", fill=TRUE, add = TRUE)
legend("topright", 
       legend = c("Muy Seguro", "Seguro", "Poco Seguro", "Inseguro"), 
       fill = c("green", "yellow", "orange", "red"),
       x = "bottomright", y = 1
)

Conclusión

Se identificó el nivel de criminalidad en los diferentes Estados de Estados Unidos y se crearon 4 divisiones en una escala del más seguro al menos seguro.