AnalyticBridge is still working on implementing an API for generic scoring of transactions, applied to:
- credit card data to detect fraud
- email data to detect spam
- click data to detect Botnets and click fraud
- loan data to detect chances of default
- customer data to detect chances of a purchase
- early terrorist activity detection
We have the technology, but we haven't had the time to implement it yet. Also, this technology relies on state-of-the-art machine learning techniques, such as hidden decision trees. Source code will be available, as well as implementation details, including source code to make the API works, and source code to create an UI that can process (in real time) data sets that are big enough to make R, JMP, Splus, Excel and similar software to crash.
Anyone interested in creating his/her own API or UI or SaaS / On Demand analytics application should contact us. We are interested in sharing our technology (at no cost) with all Analytics Service Providers.