There are 4 main types of machine learning algorithms:
- Supervised machine learning.
- Unsupervised machine learning.
- Semi-supervised machine learning.
- Reinforcement machine learning.
- Supervised machine learning.
In a supervised machine learning algorithm, the model is trained based on both the labels, and the features. During the training phase the algorithm is given the labeled data and must find the function F(x) that will allow it to determine the label based on the features given. This type of machine learning is mostly used for classification and regression.
Here is an example of data (Features, and labels) used to train a supervised machine learning algorithm:
The features (Input) are the variables used to determine the label (Output).
- Unsupervised machine learning.
In an unsupervised machine learning model, the model is trained based on features only. During the training phase, the algorithm is given data containing only features without labels. And the algorithm must find patterns and insights in this data. This type of machine learning is mostly used for clustering and association.
An example of unsupervised machine learning is K-means clustering.
- Semi-supervised machine learning.
(Labelled data is data with the features and labels included and unlabeled data is the data with the features only.)
In a Semi-supervised algorithm, the machine learning model is trained with a small quantity of labelled data and a big quantity of unlabeled data.
- Reinforcement machine learning.
In a reinforcement machine learning model, the training phase of the model is based on its own experience. It is based on trial and error. The algorithm tests a solution and receives feedback on how good it was. Based on this feedback the algorithm will adjust its future solutions and improve itself. It is a self-learning algorithm that can improve itself.
