Labeled data

As explained in Grokking Machine Learning

Data containing a designated target feature or label that a model is intended to predict.

Labeled data contains a designated feature, or label, that a model is intended to predict. A label might be a pet’s type, weight, health state, or an email’s spam status. Whether a feature counts as a label depends on the question being asked. section 2.1 section 2.5

Common label types include:

• Numerical labels, such as prices, sizes, ages, or weights, support regression models. section 2.2
• Categorical labels, such as dog/cat or spam/ham, support classification models. section 2.2
• Unlabeled data has no selected feature to predict, although the same dataset could become labeled for a different problem. section 2.1

Pet and email datasets

A pet dataset containing images plus dog/cat tags is labeled; one containing images plus weights is also labeled. An image-only dataset is unlabeled. Similarly, an email table becomes labeled when it includes a spam/ham column. section 2.1 section 2.2

Figure 2.1 Labeled data is data that comes with a tag, or label. That label can be a type or a number. Unlabeled data is data that comes with no tag. The dataset on the left is labeled, and the label is the type of pet (dog/cat). The dataset in the middle is also labeled, and the label is the weight of the pet (in pounds). The dataset on the right is unlabeled.

Compare categorical pet labels, numerical weight labels, and an image-only dataset without a target label. section 2.1

How Grokking Machine Learning covers it

How deep it goes

Defines labels, distinguishes numerical and categorical targets, and shows prediction examples without detailing specific algorithms.

Examples

Uses pet images, dog-or-cat tags, animal weights, spam-or-ham emails, and health states as concrete labeled datasets.

What it assumes

Assumes only a dataset and a prediction question, while teaching the distinction between labels and unlabeled features from examples.

What you can do after

Evaluate whether a dataset contains a target feature and identify whether its labels are numerical or categorical.

How it connects

Connects labeled data with supervised learning, regression, classification, numerical data, categorical data, and predictions.

liveBookmanning.com

Read Grokking Machine Learning on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Outlier Detection in PythonLabeled data consists of data items accompanied by target labels or annotations used for supervised learning.
• 

  liveBookmanning.com
  Natural Language Processing in Action, Second EditionA labeled dataset is a dataset whose examples are paired with target labels or annotations for supervised learning.
• 

  liveBookmanning.com
  The Art of AI Product DevelopmentLabeled data is data paired with target annotations, classifications, or outcome values for supervised learning or evaluation.
• 

  liveBookmanning.com
  Investing for ProgrammersLabeled data is data paired with known target values or categories for supervised learning.
• 

  liveBookmanning.com
  Inside AILabeled data is a dataset whose examples are paired with target labels or outcomes.

Compare how 6 books explain labeled data →

Related concepts

• supervised learningused by · Supervised learning is a machine-learning approach that learns to map inputs to target outputs from examples paired with known labels or values.
• datakind of · Data is represented information, observations, measurements, or other values that can be stored, processed, transmitted, analyzed, or interpreted.
• fine-tuningused by · Fine-tuning is adapting a pretrained machine-learning model by training it further on targeted data or objectives for a task, domain, or behavior.
• predictive analyticsused by · Predictive analytics is the use of data and statistical or computational methods to estimate future or otherwise unknown outcomes.
• reward modeluses · A reward model is a learned model that assigns scores to outputs, actions, or outcomes to represent their desirability according to preferences or evaluation criteria.
• unlabeled datacontrasts with · Unlabeled data consists of examples that lack target labels or task-specific annotations.

This explainer was generated by Manning's knowledge graph from the text of Grokking Machine Learning. The section links point to the parts of the book it draws on.