Program 14

Author

1NT23IS080 - Section B - Harsh Deep B Nair

Develop a script in R to calculate and visualize a correlation matrix for a given dataset, with color-coded cells indicating the strength and direction of correlations, using ggplot2’s geom_tile function.

Step 1: Load the required libraries

library(ggplot2)
library(tidyr)
library(dplyr)


Attaching package: 'dplyr'

The following objects are masked from 'package:stats':

    filter, lag

The following objects are masked from 'package:base':

    intersect, setdiff, setequal, union

Step 2: Use built-in dataset

We use the built-in mtcars dataset.

# Use built-in mtcars dataset
data(mtcars)

# Compute correlation matrix
cor_matrix <- cor(mtcars)

# Convert matrix to a data frame for plotting
cor_df <- as.data.frame(as.table(cor_matrix))
head(cor_df)

  Var1 Var2       Freq
1  mpg  mpg  1.0000000
2  cyl  mpg -0.8521620
3 disp  mpg -0.8475514
4   hp  mpg -0.7761684
5 drat  mpg  0.6811719
6   wt  mpg -0.8676594

Explanation:

cor(mtcars) computes pairwise correlation.
as.table() flattens the matrix into a long-format table.
The result has 3 columns: Var1, Var2, and the correlation value (Freq).

Step 3: Visualize using ggplot2::geom_tile

ggplot(cor_df, aes(x = Var1, y = Var2, fill = Freq)) +
  geom_tile(color = "white") +  # Draw tile borders
  scale_fill_gradient2(
    low = "blue", mid = "white", high = "red", 
    midpoint = 0, limit = c(-1, 1),
    name = "Correlation"
  ) +
  geom_text(aes(label = round(Freq, 2)), size = 3) +  # Show values
  theme_minimal() +
  labs(
    title = "Correlation Matrix (mtcars)",
    x = "", y = ""
  ) +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

Step	Description
`cor()`	Computes correlation values between numeric variables.
`as.table()` + `as.data.frame()`	Converts matrix into a long format suitable for plotting.
`ggplot()`	Initializes the plot using long-form data.
`geom_tile()`	Creates color-coded tiles based on correlation values.
`scale_fill_gradient2()`	Applies a diverging color scale: red (strong +ve), blue (strong -ve), white (neutral).
`geom_text()`	Adds correlation values as text in each cell.
`theme_minimal()`	Cleans up the plot visually.
`axis.text.x`	Tilts x-axis labels for better readability.