• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

Book chapter

Multi-Perspective Process Mining with Embedding Configurations into DB-based Event Logs

Process mining is a research discipline that offers methods and tools for analyzing various processes. There are a variety of process mining techniques that have in common the use of an event log as a starting point for research. In most cases, it is a flat event log (for example, in the form of a text file) containing prepared information about events. Most information systems that work with large data use the technology of relational database management systems (RDBMS) for their effective storage and processing. Recently, there has been a trend towards greater integration of RDBMSs with process mining tools. With the direct interaction of a process mining tool with a database, it becomes possible to transfer part of the “costly” data preparation operations directly to the RDBMS level. This work represents an approach in which an arbitrary database is considered as a direct data source for process mining; that is, data are extracted without using intermediate  at logs and processed directly by process mining algorithms. An approach is proposed for translating event logs represented using RDBs into their abstract representation. There is described a novel method for embedding translation schemes inside a database in the form of so-called configurations, each of which corresponds to one data perspective/process view. This allows getting “charged” self-described DB event logs and switching between different embedded perspectives without rebuilding the logs.