The BPA EyeQTM archtecture has the flexibility to grow with your need. As you ask a question (see examples on the figure), it will ask for right data, answer your questions with
confidence levels, and possible savings or incremental revenue related to decisions. For example, the network health predictor working in a telecom network
can not only predict a possible failures, but also it can estimate amount being saved because of the related decisions.
Depending on the questions that you want BPA EyeQ to answer, it may need historic and/or streaming data.
BPA interfaces uses open standard protocols (like PMML - Predictive Model Markup Language) so that you are free to choose any vendor in the marketplace keeping your cost of ownership low.
You are free to use SAS or SPSS or R working together with our platform.
Client Business logic is fed to the system at different points via XML feed. BPA's patent pending Application Based learning Component Logic (ABLCL) and Performance Based Adaptive Component
Logic (PBACL) determines and connect the best machine learning and statistical components to find answers to customer questions satisfying the performance requirements of real-time or
At the back end of the BPA EyeQ, we have libraries of machine learning and statistical analysis. It also contains our profound experience in the art of predictive modeling.