With the arrival of cloud computing and multi-core machines - we have enough compute capacity at our disposal to churn large volumes of data and dig out the hidden patterns contained in these mountains of data.
These algorithms use all the collected “past” data to learn patterns and predict results or insights that help us make better decisions backed by actual analysis.
Credit scoring, which helps the banks to decide whether to grant the loans to a particular customer or not - based on their credit history, historical loan applications, customers’ data and so on
Or the latest technological revolution from right from science fiction movies - the self-driving cars, which use Computer vision, image processing, and machine learning algorithms to learn from actual drivers’ behavior.
In all these examples, machine learning is used to build models from historical data, to forecast the future events with an acceptable level of reliability.
This concept is known as Predictive analytics.