Digital Silhouettes is the trademarked name that Predictive Networks has given to user profiles that are established through gathered click stream data and artificial intelligence (AI) processes. The profile, or cybersignature, is built from a mathematical analysis of an individual's interests as well as their keyboard and mouse activity.
By submitting your email address, you agree to receive emails regarding relevant topic offers from TechTarget and its partners. You can withdraw your consent at any time. Contact TechTarget at 275 Grove Street, Newton, MA.
A "fusion" algorithm composed of clickstream, keystroke, and mouse or pointer behavior is used to track recurring patterns. By examining the patterns over a long period of time, Digital Silhouettes is able to indentify a specific user (as opposed to device) and assign the user to one of over 140 demographic and content-related categories.
The selected demographics fall into the six major categories of gender, age, income, education, and race - all of which break down to subcategories. There are more than 90 content affinity subcategories, such as golf, pets, and car accessories, for example. Every time a user visits a Web site that is listed in an extensive Predictive Networks database, demographic and content characterization congruent with that site are added to the user's Digital Silhouette.
The more Web sites the user visits, and the longer the user is monitored, the more refined the Digital Silhouette will become. Useful statistics include average double-click intervals, ratio of double to single clicks, average mouse velocity and acceleration, and ratio of mouse to keyboard activity. Once the profile reaches a level of mathematical accuracy, participating content providers can target their marketing messages to individual Digital Silhouettes.
Predictive Networks claims that Digital Silhouettes are identified by randomly assigned and anonymous ID numbers. The company insists that Personally Identifiable Information (PII) such as names, addresses, and other private information is not known to Predictive Networks. Because of the volume and sensitivity of data gathered, however, privacy issues have been raised about Digital Silhouettes.