Evaluating Provenance-based Trust for Scientific Workflows

Provenance is the documentation concerning the origin of a result generated by a process, and provides explanations about who, how, what resources were used in a process, and the processing steps that occurred to produce the result. Such provenance information is important to improve a scientist’s ability to judge and place certain amount of trust on the generated data. We illustrate how provenance information associated with a workflow can be used to evaluate trust. This work is based on several use cases from a Bio-Diversity application. We also propose a simple architecture to illustrate our trust framework.