Correlation matrices can be overwhelming to interpret just by looking at numbers. Heatmaps provide a visual way to see the strength and direction of relationships between variables.

Why Use a Correlation Heatmap?

• Makes it easier to spot strong or weak correlations visually;
• Helps identify multicollinearity in your data;
• Uses color to communicate positive, negative, or neutral relationships;
• Especially useful when dealing with many numeric variables.

Creating the Correlation Matrix

# Select numeric columns
numeric_df <- df[, c("selling_price", "km_driven", "max_power", "mileage", "engine")]
# Compute correlation matrix
cor_matrix <- cor(numeric_df, use = "complete.obs")
View(cor_matrix)

Visualizing with ggcorrplot

ggcorrplot(cor_matrix, 
           method = "square", 
           type = "full", 
           lab = TRUE, 
           lab_size = 5, 
           colors = c("red", "white", "forestgreen"),
           title = "Correlation Heatmap",
           ggtheme = ggplot2::theme_light())

• method = "square" makes each cell a square block;
• lab = TRUE overlays the correlation values on each block;
• colors indicate direction: red (negative), white (neutral), green (positive);
• theme_light() gives the plot a clean, minimal style

Summary

• Use cor() to calculate relationships, and ggcorrplot() to visualize them;

• Color-coded matrices help you quickly grasp complex correlation patterns;

• Always clean and convert your numeric columns before running correlation analysis.