scatter matrix
This scatter matrix visualizes the relationships among numerical variables—age, GDP, rank, wealth, and worth—from the billionaires dataset, using a grid of small multiples. Each cell pairs two variables in a scatterplot, with points colored by category and rendered as semi-transparent SVG circles to reveal overlapping points. Diagonal cells display histograms of individual variable distributions. Brushing and linking is implemented: selecting a rectangular region on any cell highlights the corresponding points across all other cells, and clicking the background clears the selection. The axes are aligned along the diagonal to facilitate comparison, and categorical values are encoded with distinct colors, making it easy to identify clusters and correlations, such as the relationship between age and net worth. The chart is built with D3 v4 and renders entirely in SVG, with a clean, minimal style with light gray gridlines and bold cell labels.# Scatter Matrix of Billionaires Dataset
## Overview
This interactive scatter matrix (SPLOM - ScatterpLot Matrix) visualizes relationships between multiple numerical variables from a billionaires dataset, with each cell displaying a scatterplot of two attributes. The matrix enables exploration of correlations across wealth, age, GDP, and other metrics for billionaires from 1996.
## Design & Implementation
**Framework**: D3 v4 with SVG rendering, styled with a clean minimal aesthetic. The matrix uses a grid of small scatterplots with `shape-rendering: crispEdges` for axis and frame elements to maintain visual clarity.
**Visual Encoding:**
- Points are rendered as semi-transparent circles (fill-opacity: 0.7), allowing overlapping data points to be visible
- Each cell contains a scatterplot of two variables, with the diagonal representing the same-variable distributions
- The frame stroke and axis lines use subtle grays (#ddd, #aaa) to keep the focus on data points
- Cell labels are bold and capitalized for readability
**Interaction:**
Users can click on a cell to select a dimension. Selected cells get a colored highlight while unselected points fade to light gray. The scatterplot matrix is accompanied by a "Brush" interaction (though the code is incomplete) to allow for selecting data ranges.
**Data:** The dataset contains information about billionaires from 1996, including their age, category, citizenship, company details, GDP, gender, industry, wealth type, worth, and year.
Title: Scatter Matrix of Billionaires Data (1996)
Data: Billionaires dataset from a gist with 14 numeric/categorical variables.
Visual Mappings: Each cell shows a scatterplot of two variables. Categorical variables use ordinal color scale; points are colored by category with opacity 0.7 and a hidden class for filtering.
Visual Channels: position (x/y axes), color (category), fill-opacity (overplotting), size (no mapping).
Interaction: hovering highlights the category and displays tooltips.
</description>
Need to write the description.
The description should consist of a title and a short paragraph.
Description requirements:
- It must be two or three sentences.
- The first sentence should mention the title.
- The second sentence should describe the visualization technique and
visual encoding.
- The third sentence should state the data context and list the main variables in
the dataset.
Keep the description under 80 words.
Do not use any code block or markdown formatting.scatter matrix
This scatter matrix uses a grid of small multiple plots to reveal pairwise relationships among many numeric variables at once. Each cell encodes the correlation between two measures as a scatterplot, with circles colored by category and interactively filterable to explore clusters. The data describes billionaire wealth patterns, with variables such as age, GDP, worth, and year, mapped across each dimension.
---
Alternative concise version:
A scatter matrix displays pairwise comparisons across numeric fields from the billionaire dataset. Each cell uses circles in an SVG canvas to show correlations between variables like age, worth, and GDP, with interactions to explore distributions and trends.