Home▸Data Science▸Outlier Detection Lab 2D: Falling Dot Histogram

Outlier Detection Lab 2D: Falling Dot Histogram

Interactive 2D outlier lab: dots fall under gravity and stack into a live histogram, a linked order-statistic curve plots sorted value vs rank, and dragging the fence lines directly on the canvas re-classifies outliers by Tukey's IQR rule in real time.

Data Science2DEasy60 FPS📱 Mobile-adapted⇄ 3D version

Details

CategoryData Science

RenderingCanvas 2D

DifficultyEasy

Target FPS60

ControlsSliders for sample size and outlier injection rate regenerate the dataset and replay the drop, a fence-multiplier (k) slider re-classifies outliers live, a Regenerate Data button reseeds the random draw, a Replay Fall button re-runs the drop animation on the same data, and either orange fence line can be dragged directly on the histogram to set k by hand.

data sciencestatisticshistogramoutliersIQRexploratory data analysis

Related simulations

Anscombe's Quartet in 2D: Four Datasets, One Set of StatisticsInteractive 2D scatter-plot gallery of Anscombe's Quartet: four datasets sharing nearly…

Curse of Dimensionality: Distance Histogram & Volume CurveA 2D pairwise-distance histogram and an analytic ball/cube volume-ratio curve plotted…

Anscombe's Quartet: Why You Must Plot Your DataAn interactive 3D gallery of Anscombe's Quartet: four datasets sharing an almost identical… ⊞ All 144 in Data Science →

Sample points fall under gravity and stack into bins to build a live histogram, while a linked panel plots every sorted value against its rank — the empirical quantile function. Drag the orange fence lines straight on the canvas (or use the slider) and watch points flip red the instant they cross Tukey's 1.5×IQR rule, the same rule data scientists reach for first during exploratory data analysis.

⚙ Under the hood

Sample points fall under gravity and stack into bins to build a live 2D histogram, while a linked panel plots every sorted value against its rank as the empirical quantile function. Drag the orange fence lines directly on the canvas and watch points flip red the instant they cross Tukey's 1.5×IQR rule.

data sciencestatisticshistogramoutliersIQRexploratory data analysis

2D · HTML5 Canvas 2D · 60 FPS target · runs fully client-side, no install