experiment-package-clustering

In this repository are stored all of data used and obtained through the experiments

AGM

Here is presented the solutions for the optimization of AGM PLA

AGM

BeT

Here is presented the solutions for the optimization of BeT PLA

BeT

K-Means++

Here is presented the solutions to the K-Means++

AGM Solutions

AGM KMEANS

BeT Solutions

BeT KMEANS

DBSCAN

Here is presented the solutions to the DBSCAN

AGM Solutions

BeT Solutions

AGM Solutions

R Scripts

The script used to generate the charts are presented bellow

// Do not need to normalize because it is betwen 1-5, according the Likert Scale
data <- c(3,4,2,3,4,3,4,5,5,4,4,3,4,4,4,4,3,4)
dm1 <- c(3,4,2,3)
dm2 <- c(4,3,4,4)
dm3 <- c(5,5,4,4)
dm4 <- c(3,4,4,4,4,3,4)

// Standard Deviation
sd(data)
sd(dm1)
sd(dm2)
sd(dm3)
sd(dm4)

// Variance
var(data)
var(dm1)
var(dm2)
var(dm3)
var(dm4)

// Mean
mean(data)
mean(dm1)
mean(dm2)
mean(dm3)
mean(dm4)

// Median
median(data)
median(dm1)
median(dm2)
median(dm3)
median(dm4)


// It uses the shapiro-wik to verify if the data are in normal distribution
shapiro.test(dados)

// Not normal distribution uses Mann-Whitney
wilcox.test(mm_cros,mm_sem) 

---------------------------------------------------------------
// data:  mm_cros and mm_sem
// W = 102, p-value = 0.02229
// alternative hypothesis: true location shift is not equal to 0

// In case p-value lower than 0.05, reject null hypothesis
// If greater, accepts, that is, do not have statistic difference
 
// Boxplot:

boxplot(mm_cros, mm_sem, names=c("crossover","sem crossover"), xlab="mm")
cliff.delta(mm_cros, mm_sem)

The personal informations about DMs are not available. Any question, contact us. Best Regards.

AGM

BeT

K-Means++

AGM Solutions

BeT Solutions

DBSCAN

AGM Solutions

BeT Solutions

R Scripts

Analysis (PT-Br)

DM1 - FM

DM2 - ACLASS

DM3 - TRADEOFF

DM4 - COE

All PLA alternativas