Min-max scaling is a feature-scaling technique that linearly maps values between a specified minimum and maximum to a normalized range, commonly 0 to 1.
Min-max normalization scales each feature into a shared range, commonly 0 to 1. This prevents large raw numbers from making an artificial neural network treat one feature as more important. section 9.3
Apply the technique by:
Choosing a minimum and maximum for each feature, such as 0–120 km/h for speed or 0–360° for field of vision. section 9.3
Car-collision data
A speed of 65 km/h becomes 0.542 when scaled between 0 and 120. Terrain quality 5/10 becomes 0.5, while 400,000 km of experience becomes 1.000.
How Grokking AI Algorithms, Second Edition covers it
How deep it goes
Explains feature-specific ranges, gives the scaling formula, and demonstrates a row-by-row Python implementation.
Examples
Uses speed, terrain quality, field of vision, driving experience, and collision labels from a car-collision dataset.
What it assumes
Assumes basic feature data and classification context while explaining from zero why raw feature magnitudes can mislead an ANN.
What you can do after
Apply min-max scaling by selecting feature bounds and transforming each dataset value with a short Python routine.
How it connects
Connects feature scaling and data preprocessing with artificial neural networks and classification.
normalizationused by · Normalization is the process of transforming a value, representation, or data structure to satisfy a specified standard, scale, or structural constraint.
predict_probaused by · predict_proba is a classifier method that returns estimated probabilities for each possible class for supplied input samples.
feature scalingimplements · Feature scaling is the transformation of numerical features to comparable magnitudes, ranges, or distributions for computation or model fitting.
anomaly scoreuses · An anomaly score is a numerical measure of how anomalous an observation or event is relative to a reference or expected pattern.
scikit-learnpart of · scikit-learn is an open-source Python library providing tools for classical machine learning, including algorithms, preprocessing, model selection, and evaluation.
RobustScaleralternative to · RobustScaler is a scikit-learn transformer that scales features using robust statistics such as the median and interquartile range.
This explainer was generated by Manning's knowledge graph from the text of Grokking AI Algorithms, Second Edition. The section links point to the parts of the book it draws on.
