Unsupervised learning is a type of machine learning in which a machine learning model is trained to find patterns or structures in the input data without the help of labels or previous answers.

Unlike supervised training, in which the model is trained with labelled examples that indicate the correct answers, in unsupervised training the model must find patterns on its own. This approach is useful in situations where labels are not available or when the correct answers are not known.

There are several unsupervised training algorithms used in machine learning, such as clustering, dimensionality reduction and anomaly detection.

Clustering is a grouping algorithm that divides data into groups or clusters based on the similarity between them, which allows patterns or structures in the data to be identified. A typical example of a clustering model is Self-Organising Maps (SOM).

Dimensionality reduction is a process that reduces the number of variables or features in the data, which can help visualise and analyse the data more efficiently.

Anomaly detection is a process that looks for outliers or exceptional values in the data, which can be useful in detecting fraud or errors in the data.