Trustev is a new approach to stopping fraud online, while letting through good customers. It analyzes thousands of datapoints in real time to make an instant decision on each transaction: is it real or fraudulent?

With enterprise customers in ecommerce, financial services, virtual goods and other industries, Trustev is the most effective way to analyze online sales, signups and logins, without the high costs of manual review -- and with much greater accuracy.

The system is the first "hybrid" approach which combines machine learning with a decision matrix, for the most accurate possible decisions. The analysis includes:

--- Behavior --- How did a customer behave on the website? Was it like a "normal" customer or not?

--- Device --- Every customer on a site is analyzed across whole visit, not just at checkout.

--- IP --- Deep analysis of network, proxy, geo-location.

--- Email -- Syntax, ping, known fraud.

--- Phone --- Cross-reference phone account number, tenure, network

--- History --- Full customer history integration and analysis

--- Social --- Does this look like a real person, or not?

--- Machine learning --- 1000s of datapoints fed into a feedback loop, to improve continuously over time