Base120 · Decomposition

Signal Separation

DE · Decomposition

Distinguish meaningful patterns from random variation or confounding factors

Model code
DE9
Transformation
DE (Decomposition) — Break problems into factors, layers, and independent subproblems.
Priority
9 of 20
Framework
HUMMBL Base120 — 120 mental models across 6 transformations

When to use

"Use when data contains both meaningful patterns and random variation. Separate signal (the pattern you care about) from noise (random variation or confounding factors) before drawing conclusions."

Example

"A/B test results show a 3% lift. Is it signal (the change caused the lift) or noise (random variation)? Statistical significance testing separates the two — if the result is within the range of random variation, it is noise."

Common misuse

"Treating all variation as signal. Overfitting occurs when noise is treated as pattern — the model learns the random variation in the training data and fails to generalize."

Also known as

Signal ExtractionNoise Filtering

Related models

DE5DE14RE11