AWS Certified Machine Learning – SpecialtyExploratory Data AnalysisMedium
A data scientist is analyzing a dataset of customer demographics, including 'age' and 'income'. They want to visualize the joint distribution of these two continuous variables, specifically to identify areas of high customer density without making assumptions about the underlying distribution shape. Which visualization technique is most suitable?
- A2D Kernel Density Estimate (KDE) plot
- BBox Plot
- CHistogram
- DBar Chart
Show answer & explanationAnswer & explanation
Correct answer: A. 2D Kernel Density Estimate (KDE) plot
A 2D Kernel Density Estimate (KDE) plot is designed to visualize the joint probability density function of two continuous variables. It creates a smooth surface representing data density, effectively highlighting areas of high concentration without assuming a specific distribution shape, which is perfect for identifying customer density 'hotspots'.
Why the other options are wrong
- B. A box plot summarizes the distribution of a single continuous variable and is not suitable for visualizing joint distributions or density hotspots.
- C. A histogram shows the distribution of a single continuous variable, not the joint distribution of two.
- D. A bar chart is used for categorical data or discrete counts and is not appropriate for visualizing the joint distribution of two continuous variables.
2D Kernel Density Estimate (KDE)
A non-parametric method for estimating the probability density function of two continuous variables, creating a smooth surface that highlights areas of high data concentration.
- Visualizes joint distribution of two continuous variables.
- Non-parametric; makes no assumptions about distribution shape.
- Useful for identifying clusters or 'hotspots' in data.
Memory trick: KDE: 'Kids Drawing Everywhere' - smooth density contours.