Table 1.

Types of machine learning methods.

Learning type	Supervised	Unsupervised	Semi-supervised
Type of data	Data points have labels.	Data points do not have corresponding labels.	A subset of the data points is labeled.
Learning process	Analyzing the training data to learn a function that can be used for predicting the labels of new examples.	Modeling the structure or the distribution of the data in order to find patterns and gain new insights from the data.	Utilizing unlabeled data with labeled data to learn better models.
Applications	Fraud detection, detecting spam emails, predicting real estate prices.	Clustering customers' data and market segmentation, learning rule associations, image segmentation, gene clustering.	When it is expensive to annotate every data point (e.g., using humans), this type of learning is suitable. Examples: web content classification, medical predictions.

Firstly, the nature of the data is stated, then the objective of the type of learning is discussed, and finally some real-world examples are mentioned.