Supervised Learning

What is Supervised Learning?

Supervised learning is part of machine learning, and it deals with labelled data. The operation behind supervised learning adopts training of algorithms (classification or regression) through the use datasets. Labelled data refers to information, containing tags (labels) of correct outputs (answers). This type of machine learning technique uses datasets, called training data, which through the use of known answers teaches and enables to be able to predict accurate outcomes from new unseen data.

Types of Supervised Learning

There are two main types of supervised machine learning algorithms, classification and regression – with some algorithms able to deal with both problems.

Classification

Classification algorithms tackle categorical problems. This method learns to classify inputs into the correct class, such as ‘Yes’ or ‘No’ and ‘True’ or ‘False’. Questions with two classes refer to binary classification and questions with more than two classes to multiclassification

Example of classification
The goal of the model is to identify new incoming emails as spam or not. The model is supplied with a training data of thousands of emails, with the correct label of spam or not for each. Through the training data, the model learns the characteristics of a spam email and subsequently how to identify one.

supervised machine learning classification diagram

Regression

Regression algorithms tackle continues variables. This technique learns the relationship between variables (dependent and independent) and predicts an output, such as predicting real estate price. Regression in machine learning deals with numerical values.

Example of regression
The objective of the model is to predict unknown property prices. Already identified thousands of properties, together with their features such as number of bedrooms, location, and the corresponding purchase price are supplied to the model. Then through training, the model learns how to predict the price of a property which has not been yet valuated.

supervised machine learning regression diagram

Applications of Supervised Learning

Supervised learning has very large implementation nowadays, with many applications in multiple industries. The following are only a small part of the many uses of this type of machine learning.

Image and object recognition in social media
Email spam detection
Risk assessment in finance
Valuation and investment prediction
Predictive targeting in marketing and advertising

Next: Unsupervised Machine Learning

by AICorr Team

We are proud to offer our extensive knowledge to you, for free. The AICorr Team puts a lot of effort in researching, testing, and writing the content within the platform (aicorr.com). We hope that you learn and progress forward.