AI Glossaries

A validation dataset is a portion of data held back from training and used to tune a machine learning model, compare options and decide when to stop training, before the final test on unseen data.

A training dataset is the portion of data used to teach a machine learning model. The model studies these examples and adjusts its internal parameters to learn the patterns that connect inputs to outputs.

Feature engineering is the process of turning raw data into input variables, called features, that help a machine learning model learn better, for example converting timestamps into day of week or combining columns into ratios.

Clustering is an unsupervised machine learning technique that groups similar data points together without using labels, helping reveal natural segments such as customer types, document topics or unusual behaviour.

Regression in machine learning is a supervised learning task where a model predicts a continuous number, such as a price, a temperature or a delivery time, based on patterns learned from labelled examples.

Classification in machine learning is a supervised learning task where a model learns to assign inputs to predefined categories, such as spam or not spam, by learning decision boundaries from labelled examples.

Self-supervised learning is a machine learning approach where a model creates its own training labels from raw data, for example by predicting a hidden word or the next word in a sentence. It is the main method behind large language models.

Semi-supervised learning is a machine learning approach that combines a small amount of labelled data with a large amount of unlabelled data, reducing labelling costs while still achieving strong accuracy.

Reinforcement learning (RL) is a type of machine learning where an agent learns to make decisions by trying actions in an environment and receiving rewards or penalties, gradually learning the strategy that earns the most reward over time.

Unsupervised learning is a type of machine learning that finds patterns, groups and structure in data without labelled answers, used for tasks like customer segmentation, anomaly detection and simplifying complex data.

Supervised learning is a type of machine learning where a model learns from labelled examples, meaning each input comes with the correct answer, so it can predict the answer for new, unlabelled inputs.

Machine learning (ML) is a branch of artificial intelligence in which computers learn patterns from data and improve at a task through experience, instead of being programmed with every rule by hand.