Machine Learning Algorithms

Machine Learning Algorithms

Machine Learning Algorithms

Short description explaining what the learner will understand after completing this study map.

9 sections Interactive study map Knowledge check
LEARN
01

What Is a Machine Learning Algorithm?

A machine learning algorithm is a method used to learn patterns or relationships from data. Different algorithms are suited to different kinds of problems, such as predicting numerical values, assigning categories, or grouping similar observations.

Key Point

Core idea: first identify the type of problem. Regression predicts numerical values, classification assigns categories, and clustering discovers groups based on similarity.

02

Essential Algorithms

Tap each card to review common machine learning algorithms and their typical purpose.

03

Three Main Algorithm Tasks

Common algorithms can be grouped by the kind of result they are intended to produce. Regression estimates numerical values, classification chooses categories, and clustering groups observations without requiring predefined class labels.

04

Regression vs Classification

Regression and classification are both commonly supervised tasks, but their outputs differ. Regression predicts a continuous numerical value, while classification predicts a discrete class or category.

Regression

Use regression when the desired output is numerical, such as estimating a price, temperature, demand level, or another continuous quantity.

Classification

Use classification when the desired output is a category, such as yes/no, approved/rejected, or one class from several possible classes.

05

Algorithm Families at a Glance

Several algorithms can address the same broad task in different ways. Understanding their basic behavior helps you recognize which family an algorithm belongs to and what kind of output it is designed to produce.

REGRESSION → Linear / Polynomial

CLASSIFICATION → KNN / Decision Tree / Logistic Regression

CLUSTERING → K-means / Hierarchical
06

How the Algorithms Differ

Linear regression fits a straight-line relationship; polynomial regression can represent curved relationships. K-nearest neighbors uses nearby examples, decision trees follow branching decisions, logistic regression estimates class probability, K-means forms clusters around centers, and hierarchical clustering builds nested groups.

07

Algorithm Explorer

Explore the major algorithm families and recognize the type of problem each is designed to address.

08

Key Takeaways

Focus on the problem and output: numerical prediction → regression; known categories → classification; unlabeled grouping → clustering.

Remember

Core pattern: regression predicts numerical values, classification predicts categories, and clustering groups similar data. The algorithm should match the problem being solved.

Common Mistake

Common mistake: despite its name, logistic regression is commonly used for classification. Do not assume that every algorithm containing the word “regression” produces a continuous numerical prediction.

09

Knowledge Check

These questions test whether you can identify common machine learning algorithms and match them to regression, classification, or clustering tasks.

Score: 0 / 0
EXIN BCS MACHINE LEARNING AWARD · AIMLA 1.1 · DEFINING MACHINE LEARNING