A knowledge discovery service system for provenance exploration

Scientific research has moved from an isolated environment into a collaborated culture due to the data explosion and the experiment complexity. Several scientific workflow systems have been developed to support scientists to conduct the scientific analysis and create knowledge. They provide the ability to automatically record provenance, the process that led to a particular data, which can be used by
other scientists for better understanding, re-using and verifying a data product to generate new knowledge. However, most of the existing scientific workflow systems
provide poor query capability for users to effectively find relevant provenance information. It is not a trivial task due to the complexity of the workflow and the size of the provenance graphs. In this work, we propose a system architecture for provenance exploration in which querying, navigation and visualization methods are combined together. It enables effective and efficient knowledge discovery
among various provenance information generated by different workflow runs.