Mechanism and techniques for discovery and prediction of long-term high-level nearly periodical business processes from low-level logs

摘要

A method, apparatus and computer program product for providing discovery and prediction of long-term high-level nearly periodical business processes from low-level logs is presented. Query activity vectors are extracted from database logs. Groups of queries having similar behavior for a set of business processes are identified by clustering of the query activity vectors. Periodic structure is discovered in the business processes and their future activity is predicted. The predictions of future activity are provided to a storage system.