Technology Overview

Increased Precision to Provide New Insights

Only Inomaly’s novel solution gives its customers the power to find a needle in an electronic haystack – when you need knowledge, not confidence.

Inomaly has addressed the need for systematically identifying critical anomalies in large and complex data sets. Our patent-pending technology identifies important relationships in data sets that other analytics tools are not designed to detect because they are constrained by a fundamental problem in mathematics solved by Professor Schlottmann. These other tools use methods that group data into like-minded bunches in order to make predictions about how the dataset, as a whole or a subset thereof, will act on average. Such an approach is primarily useful for making broad clinical, policy, or business decisions if the decision-maker is looking for a certain degree of confidence in a particular outcome. By design relationships that do not occur with sufficient frequency across the data are averaged out or "swamped" into non-existence by such trend detection methods – leaving end users in the dark with respect to significant risks or potential opportunities. Inomaly's break-through technology helps end users systematically detect these infrequent data points and relationships with precision because it does not rely on frequency across the data to detect important relationships therein. By solving the mathematical problem of detecting low frequency events in large data sets, Inomaly's technology closes the loop on data analysis by giving its customers insight into trends and the less frequent relationship that may have big consequences if overlooked.

Growth in Data Warehouses Necessitates New Tools for Finding Critical Anomalies

As data warehouses have grown exponentially in size and complexity in recent years, the information needed has become increasingly rare or anomalous. This hard-to-find information may represent key opportunities for increasing profit or decreasing losses. Enabling organizations to systematically find critical anomalies and augment existing predictive analysis is the next major challenge in the analytics community of today's information saturated environment.

Mortgage Industry Example

In 2006, Professor Schlottmann applied a prototype version of Inomaly's current analytics engine to a major US lending institution's portfolio of approximately 1 million loans. He successfully identified approximately $700 million in "below acceptance grade" risk in a portfolio that was missed by the industry-leading products already in use by the bank and its competitors.