REAL-TIME MACHINE LEARNING ARCHITECTURE EMBEDDED IN IN-MEMORY TRANSACTION DATABASE SYSTEM

摘要

The present invention provides a method for embedding a machine learning model in an in-memory transaction database system and updating it in real time and performing a real-time transaction process on a dynamic learning model at the same time. A dynamic machine learning model is embedded to update the machine learning model at every transaction. This dynamic machine learning model can be used for a transaction process.;COPYRIGHT KIPO 2017