KEfED Modeling Approach

The basic idea is that we model the experimental process that gives rise to data, all the while tracking how metadata propagates through the experimental workflow. This provides a generative model for the data that is then obtained from the experiment. Therefore, the KEfED modeling methodology provides a framework for understanding the design of data generated from experiment.

Basic Process

1. Model the protocol in terms of processes, subprocesses and entities.
2. Model the independent variables that parameterize processes and entities.
3. Reconstruct the provenance of these parameters through the workflow to provide a data context for each measurement in the final data

We adopt a diagrammatic methodology to assist the process.