Predictive technology is a body of tools capable of discovering and analyzing patterns in data so that past behavior can be used to forecast likely future behavior. Predictive technologies, which include data mining , neural networks, and system modeling and simulation, have been applied to the study of weather systems, traffic patterns, stock markets, epidemiology, consumer behavior, and terrorist activity, and many other areas of study where there can be a significant number of variables.
Predictive technology is increasingly used for marketing purposes. Many retailers, for example, collect barcode data at their point-of-sale terminals. The raw data can be processed to get predictions about consumer behavior for future campaigns. Amazon.com was one of the earliest online retailers to use predictive technology. The Amazon Web site compiles lists of merchandise that visitors look at and displays lists of items that customers with similar interests have bought.
Another example of predictive technology is DARPA's proposed Total Information Awareness (TIA) system. The TIA project's ultimate goal is to amass enormous volumes of information about individuals (this collection of data is sometimes called an individual's information signature) from all available sources, to process that data and use predictive technology to identify and avert potential terrorist activity.
Back to: Glossary