Predictive analytics is the umbrella term for analyzing patterns found in data to predict future behavior or results. It includes techniques and algorithms found in statistics, machine learning, artificial intelligence, and data mining. Popular predictive analytics techniques include regression, classification, clustering, and decision trees.
When using machine learning, predictive analytics models can be created with supervised or unsupervised learning techniques. Supervised learning uses a known set of data and measures the hypothesis of the analyst against specific target attributes. Unsupervised learning lets the system analyze the data and discover previously unknown patterns mathematically on its own.
Popular use cases for predictive analytics include risk analysis, fraud detection, customer behavior, and anticipating necessary maintenance.