EDA Visualizations — SegmentIQ

Age Distribution

Most customers are between 25–45 years old. The distribution is slightly right-skewed, indicating the mall attracts a younger-to-middle-age demographic.

Annual Income Distribution

Income ranges from $15k to $137k. The peak is around $60k–$70k, suggesting a predominantly middle-income customer base.

Spending Score Distribution

Spending scores are fairly uniformly distributed with slight peaks at both extremes, suggesting two distinct spending personalities — frugal and impulsive.

Gender Distribution

The dataset has slightly more female customers than male, which may reflect shopping centre demographic trends.

Boxplots — Outlier Detection

Reading the boxplot: The box spans Q1–Q3 (IQR). Whiskers extend to 1.5×IQR. Individual points beyond whiskers are outliers. This dataset has very few outliers — the data is naturally clean.

Age vs Spending Score

Younger customers (18–35) show higher spending scores on average. There is a weak negative correlation between age and spending score.

Annual Income vs Spending Score

This is the key plot for clustering. Five distinct groups are visually apparent — this confirms K=5 is appropriate.

Correlation Heatmap

Age ↔ Income: Near-zero correlation (0.03). Age does not predict income level.

Age ↔ Spending: Weak negative (−0.33). Older customers spend slightly less on average.

Income ↔ Spending: Near zero (0.01). Income and spending are essentially independent — ideal for K-Means.

Next: Elbow Method

📊 EDA Visualizations