Clustering - Kmeans¶

In [1]:

# DAL ToolBox
# version 1.01.727

source("https://raw.githubusercontent.com/cefet-rj-dal/daltoolbox/main/jupyter.R")

#loading DAL
load_library("daltoolbox")

Loading required package: daltoolbox

Registered S3 method overwritten by 'quantmod':
  method            from
  as.zoo.data.frame zoo 


Attaching package: ‘daltoolbox’


The following object is masked from ‘package:base’:

    transform

In [2]:

#load dataset
data(iris)

General function to test clustering methods¶

In [3]:

# setup clustering
model <- cluster_kmeans(k=3)

In [4]:

# build model
model <- fit(model, iris[,1:4])
clu <- cluster(model, iris[,1:4])
table(clu)

clu
 1  2  3 
96 21 33

In [5]:

# evaluate model using external metric
eval <- evaluate(model, clu, iris$Species)
eval

$clusters_entropy

A tibble: 3 × 4
x	ce	qtd	ceg
<fct>	<dbl>	<int>	<dbl>
1	0.9987473	96	0.63919827
2	0.7024666	21	0.09834532
3	0.0000000	33	0.00000000

$clustering_entropy

0.737543587859802

$data_entropy

1.58496250072116

Influence of normalization in clustering¶

In [6]:

iris_minmax <- transform(fit(minmax(), iris), iris)
model <- fit(model, iris_minmax[,1:4])
clu <- cluster(model, iris_minmax[,1:4])
table(clu)

clu
 1  2  3 
21 96 33

In [7]:

# evaluate model using external metric
eval <- evaluate(model, clu, iris_minmax$Species)
eval

$clusters_entropy

A tibble: 3 × 4
x	ce	qtd	ceg
<fct>	<dbl>	<int>	<dbl>
1	0.7024666	21	0.09834532
2	0.9987473	96	0.63919827
3	0.0000000	33	0.00000000

$clustering_entropy

0.737543587859802

$data_entropy

1.58496250072116