Load Dataset

The necessary libraries and the wine dataset are loaded.

library(tidyverse)
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr     1.1.4     ✔ readr     2.1.5
## ✔ forcats   1.0.0     ✔ stringr   1.5.1
## ✔ ggplot2   3.5.1     ✔ tibble    3.2.1
## ✔ lubridate 1.9.3     ✔ tidyr     1.3.1
## ✔ purrr     1.0.2     
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
data_url <- 'https://raw.githubusercontent.com/koimabrian/Datasets/refs/heads/main/Wine.csv'
Wine <- read.csv(data_url, comment = "#")  # Handle potential comments in the CSV
head(Wine)

Standardize Continuous Variables

The continuous variables (excluding observation number and country) are scaled.

data <- Wine %>% select(-Obs, -Country) %>% scale()

Create Clusters with K-Means

The k-means algorithm is applied to the standardized data, setting the number of clusters to 3.

kmeans_result <- kmeans(data, centers = 3, iter.max = 100, nstart = 100)
kmeans_result
## K-means clustering with 3 clusters of sizes 25, 43, 32
## 
## Cluster means:
##          Rating       Price    Alcohol Residual_Sugar  Sulphates          pH
## 1  0.0005649203 -0.09531108  0.7695537     0.73705964  0.7738343 -0.71133001
## 2 -0.7611830170 -0.74488647 -0.3311789    -0.48277078 -0.1169446  0.37747968
## 3  1.0223983350  1.07540297 -0.1561923     0.07289539 -0.4474138  0.04848825
## 
## Clustering vector:
##   [1] 3 2 2 1 2 2 1 1 3 3 2 2 2 2 2 3 3 2 2 2 1 1 3 3 2 2 3 1 2 3 2 3 3 1 3 2 3
##  [38] 1 3 2 2 2 3 2 1 3 2 2 3 3 1 1 2 3 2 2 3 1 2 2 1 2 3 1 3 3 3 1 2 2 2 2 2 1
##  [75] 3 2 2 1 1 2 2 3 1 1 2 3 2 1 1 1 3 2 1 3 3 3 3 2 1 2
## 
## Within cluster sum of squares by cluster:
## [1] 114.3882 156.1271 119.1354
##  (between_SS / total_SS =  34.4 %)
## 
## Available components:
## 
## [1] "cluster"      "centers"      "totss"        "withinss"     "tot.withinss"
## [6] "betweenss"    "size"         "iter"         "ifault"

Determine the Optimal Number of Clusters

To identify the optimal number of clusters, the within-cluster sum of squares (WSS), silhouette method, and gap statistic are visualized.

library(factoextra)
# WSS plot
fviz_nbclust(data, kmeans, method = "wss")

# Silhouette plot
fviz_nbclust(data, kmeans, method = "silhouette")

# Gap statistic
fviz_nbclust(data, kmeans, method = "gap_stat")

Cluster Visualization

Clusters are visualized in a biplot, and then plotted using the original variables (Rating vs Price).

# Cluster biplot
fviz_cluster(kmeans(data, centers = 3, iter.max = 100, nstart = 100), data = data)

Clusters with Original Variables

Cluster labels are added to the original dataset, and clusters are visualized using the Rating and Price variables.

clusters <- kmeans(data, centers = 3, iter.max = 100, nstart = 100)
Wine <- Wine |> mutate(cluster = clusters$cluster)

# Plot clusters using Rating and Price
Wine |> ggplot(aes(x = Rating, y = Price, col = as.factor(cluster))) + 
  geom_point() + 
  labs(title = "Wine Clusters by Rating and Price", color = "Cluster")