PriceTag HUB / AI

Artificial Intelligence

Understand artificial intelligence, its applications, technologies, tools, and impact on the world.

Knowledge Hub

Learn AI step by step

Artificial Intelligence · Machine Learning

Types of Machine Learning

Explore the main types of machine learning, understand how each approach learns from data or feedback, and discover examples of where these methods are used in real-world applications.

Explore the Types Machine Learning Guide

Understanding Learning Methods

How is machine learning classified?

Machine learning includes different approaches for teaching computer models to learn patterns and solve problems. These approaches differ in the kind of data they use, the feedback available during learning, and the objectives they are designed to achieve.

Supervised learning uses examples with known answers, while unsupervised learning searches for patterns in data without predefined labels. Semi-supervised learning combines labeled and unlabeled examples, whereas reinforcement learning focuses on actions, interaction, and rewards.

Each approach is suitable for different situations. The right choice depends on the problem, the available data, the expected result, and the resources needed to train and evaluate a model. Understanding these differences makes it easier to see how machine learning is applied across industries.

Main Categories

Four Types of Machine Learning

Learn how each approach works, what kind of information it uses, and which real-world problems it can help solve.

Supervised Learning

Supervised learning trains a model using labeled examples, where the input data is paired with a known answer. The model learns relationships between inputs and outputs to make predictions on new data.

Examples

Email spam detection, house price prediction, and image classification.

Common methods

Classification and regression.

Unsupervised Learning

Unsupervised learning works with data that does not have predefined labels. Models identify patterns, similarities, groups, or underlying structures that may be useful for understanding the data.

Examples

Customer segmentation, grouping similar documents, and discovering patterns in purchasing behavior.

Common methods

Clustering and dimensionality reduction.

Semi-Supervised Learning

Semi-supervised learning combines a smaller amount of labeled data with a larger amount of unlabeled data. It can be useful when labeling examples requires significant time, effort, or expert knowledge.

Examples

Image classification and document categorization when only a portion of the available examples has labels.

Common methods

Training approaches that use both labeled and unlabeled data.

Reinforcement Learning

Reinforcement learning trains an agent to make decisions by interacting with an environment. The agent receives rewards or penalties based on its actions and learns a strategy that aims to maximize long-term reward.

Examples

Game-playing systems, robotics, and decision-making in simulated environments.

Common methods

Reward-based learning and policy optimization.

Quick Comparison

Difference Between Machine Learning Types

The key difference is how each approach receives information during learning and what it tries to achieve.

Type	Learning Input	Main Goal
Supervised	Labeled data	Predict known outcomes
Unsupervised	Unlabeled data	Discover patterns
Semi-supervised	Labeled and unlabeled data	Learn with limited labels
Reinforcement	Rewards and interaction	Learn effective actions

Choosing the Right Method

Which Type of Machine Learning Should You Use?

The most suitable approach depends on the problem you want to solve. If you have examples with known answers and want to predict outcomes for new cases, supervised learning may be appropriate. If you want to discover groups or relationships without predefined labels, unsupervised learning may be useful.

Semi-supervised learning can help when only some of your data has labels. Reinforcement learning is worth considering when an agent must learn a sequence of decisions through interaction and feedback from its environment.

These categories are useful starting points, not rigid rules for every project. Real-world systems may combine multiple techniques depending on their requirements, data, and design.

Continue Learning

Explore More About Machine Learning

Understand the fundamentals and discover how different learning methods are applied to real-world problems.

What Is Machine Learning?Uses of Machine Learning Machine Learning Main Page All AI Topics