Standardization

As explained in Knowledge Graphs and LLMs in Action

Feature standardization transforms numerical features so that they have a specified mean and scale, commonly zero mean and unit variance.

Feature standardization transforms numerical features to zero mean and unit variance, preventing large-scale measurements from dominating model decisions. section 9.3.1

Apply standardization as a consistent training and prediction pipeline:

• Fit StandardScaler on the training feature vectors. section 9.3.1
• Transform the training vectors before fitting logistic regression. section 9.3.1
• Reuse the fitted scaler to transform test or prediction vectors. section 9.3.1

Node features on different scales

Node degree may range from 1 to 1,000 while centrality measures range from 0 to 1. Scaling both features prevents degree from overwhelming centrality in distance-based calculations. section 9.3.1

Figure 9.7 Training flow for node classification and link prediction. A critical step is the featurization of nodes and relationships. When this process is finished, the vectors can be passed to a classic algorithm. Both node classification and link prediction can be seen as classification tasks.

Use the training flow to locate standardization between node and relationship featurization and the downstream classifier. section 9.3.1

How Knowledge Graphs and LLMs in Action covers it

How deep it goes

Demonstrates fitting and reusing StandardScaler, but does not derive the transformation mathematically.

Examples

Uses node degree, centrality measures, graph embeddings, and the karate club network as concrete feature examples.

What it assumes

Assumes numerical feature vectors and a downstream machine-learning classifier.

What you can do after

Apply it via StandardScaler to align training and prediction features before fitting and evaluating a classifier.

How it connects

Connects feature scaling with embeddings, Euclidean distance, logistic regression, and node classification.

liveBookmanning.com

Read Knowledge Graphs and LLMs in Action on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Statistics Slam DunkStandardization transforms numeric variables to a common scale, commonly by subtracting the mean and dividing by the standard deviation.
• 

  liveBookmanning.com
  Machine Learning for Tabular Data: XGBoost, Deep Learning, and AIA standardized feature is a feature transformed to a normalized scale, commonly with a specified mean and standard deviation.
• 

  liveBookmanning.com
  Ensemble Methods for Machine LearningStandardization rescales numerical features so that each feature has a specified mean and dispersion, commonly zero mean and unit standard deviation.
• 

  liveBookmanning.com
  Data Without LabelsStandardization is the transformation of numeric variables to a common scale, typically by centering them and scaling by their variability.

Compare how 5 books explain standardization →

Related concepts

• standard deviationuses · Standard deviation is the nonnegative square root of variance and measures the typical dispersion of values around their mean in the original units.
• z-scoreproduces · A z-score is a signed standardized value expressing how many standard deviations an observation lies above or below a reference mean.
• StandardScalerimplemented by · StandardScaler is a scikit-learn preprocessing transformer that centers each feature using its fitted mean and scales it using its fitted standard deviation.
• feature scalingkind of · Feature scaling is the transformation of numerical features to comparable magnitudes, ranges, or distributions for computation or model fitting.
• pipelineused by · A pipeline is an ordered or coordinated sequence of connected processing stages or tasks that passes data, work, or results toward an outcome.
• preprocessingkind of · Preprocessing is the preparation or transformation of data, content, or other inputs into a form suitable for a subsequent computational operation.

This explainer was generated by Manning's knowledge graph from the text of Knowledge Graphs and LLMs in Action. The section links point to the parts of the book it draws on.