This article briefly describes the types of Machine Learning problems that the AI & Analytics Engine can tackle.
The AI & Analytics Engine currently supports three types of Machine Learning problems, namely classification, regression, and clustering.
Classification (supervised learning)
Given an observation, predict which category it belongs to, from a set of categories.
Examples:
-
Binary classification -- Is an email spam or not (yes/no)?
-
Multi-class classification -- Which genre does a movie belong to (action, comedy, drama, fantasy, romance, etc.)
Regression (supervised learning)
Given an observation, predict a numeric outcome such as house price and taxi trip duration.
Tip: Still curious? For a detailed comparison between classification and regression, see this article.
Clustering (unsupervised learning)
Partition data in the way that similar entities are grouped into the same cluster.
Examples:
-
Segment customers into groups of similar characteristics to target
-
Group items in a search engine and show similar results for a query
Tip: You can find some typical methods and applications of clustering in this article.