Home▸AI & Machine Learning▸Decision Threshold Profit Optimizer 2D

Decision Threshold Profit Optimizer 2D

2D companion to the 3D decision-threshold optimizer: the same cost-sensitive classification math — two overlapping score-distribution curves, a movable threshold and an expected-profit curve — plus a class-balance slider and a live confusion-matrix breakdown.

AI & Machine Learning2DModerate60 FPS📱 Mobile-adapted⇄ 3D version

Details

CategoryAI & Machine Learning

RenderingCanvas 2D

DifficultyModerate

Target FPS60

ControlsSliders: decision threshold (0-1, default 0.5), positive-class prevalence (5-70%, default 30%), benefit of a caught true positive ($10-300, default $100), cost of a false positive ($1-150, default $20), cost of a false negative ($1-300, default $80); Jump to Optimal Threshold button snaps the threshold to the profit-maximizing cutoff for the current settings.

machine learningbusiness analyticsclassificationdecision theorycost-sensitivechurn

Related simulations

Gradient Descent VisualiserWatch SGD, Momentum, RMSprop and Adam navigate a 3D loss landscape in real time. Compare…

K-Means Clustering — Machine Learning VisualiserVoronoi regions and centroid trails update as K-Means++ steps through blob, half-moon or…

Convolutional Neural NetworkA filter scans an 8×8 image layer by layer — watch convolution, ReLU activation and… ⊞ All 1187 in AI & Machine Learning →

This is the 2D companion to the 3D decision-threshold scene, computing the identical cost-sensitive classification math on a flat chart: overlapping positive- and negative-class score-density curves, a movable decision line and a live expected-profit curve traced across every possible threshold. It adds a class-balance (prevalence) slider and a full confusion-matrix breakdown (TP/FP/FN/TN shares) so you can see exactly how a rarer positive class, not just the cost weights, pushes the profit-maximizing cutoff away from the textbook 0.5.

⚙ Under the hood

2D companion to the 3D decision-threshold optimizer: overlapping positive/negative score-density curves, a movable decision line and a live expected-profit curve, plus a class-balance slider and a full confusion-matrix breakdown.

machine learningbusiness analyticsclassificationdecision theorycost-sensitivechurn

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