Saltar al contenido principal

Data rodeo execution

In this section is described what is been done with the data in each phase of a data rodeo.

Data rodeo execution Figure 8. Data rodeo execution

  • Factories. During the ingestion process the data is treated row by row, then, after the ingestion process the data is ready to generate a Process Mining Log per ingestor. Each Log includes a unique trace per ID. Once the Log is ready filters can be applied to each one.
  • Filters. Filters can be applied to the logs generated during the ingestion process such as split filters to cut the unique traces into several, stratification filters to calculate new Trace and Event metadata, corrective filters to correct or simplify the log according to the health professional criteria or discretizer filters to make more understandable numerical values.
  • Log processors. This phase is focused on applying the discovery process mining algorithms in order to identify the final log.
  • TPA Processors. In this phase, trace data can be used to calculate statistics, event and trace metadata and/or abstractions.
  • Renders. In this phase, the process is presented including techniques of conformance and/or enhancement that will help to highlight the especifities of the model.