In Ammlakh/Lab-7-Group: Basic clustering implemented by hand

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

clust431

The goal of clust431 is to do k-means clustering and hierarchical clustering.

Installation

You can install the released version of clust431 from CRAN with:

install.packages("clust431")

Example for k_means()

This is a basic example which compares the original kmeans function to the one made for lab 7

library(clust431)
library(tidyverse)

iris2 <- iris %>%
    select(Sepal.Length, Sepal.Width)
k_means(iris2, 3, PCA = F)

The chunk below is comparing the results of my function from above to the built-in function.

actual <- kmeans(iris2, 3)
actual
actual$totss - actual$betweenss
actual$totss

As we can see, we get the same SSE, clustering vectors, and cluster means. Our function gives SSE straight up, but there is more calculations needed. That being said, the built-in function has much more information.

Example for hier_clust()

This is a basic example which compares the actual hier_clust function to the one made for lab 7

iris3 <- iris %>%
  select(-c(Species))
b = hier_clust(iris3, k=150, method = 'euclidean')
head(b, 6)
a = hclust(dist(iris3))
head(a$merge, 6)

hier_clust(iris3, k=3, method = "manhattan")

hier_clust(iris3, k=4, method = "euclidean")

Ammlakh/Lab-7-Group documentation built on June 4, 2022, 12:01 p.m.