A Query Language of Data Provenance Based on Dependency View for Process Analysis

Abstract

Abstract-For the scale of data in process keep increasing, data provenance also becomes large and constantly growing, which brings challenges to the efficiency of provenance tracking in process analysis. This paper proposes a kind of dependency view to extract a global data provenance description of the data process instance, and then defines a contextual query language based on dependency view to implement an efficient provenance query mechanism for process analysis. The elements of the language are based on a set of dependency view query operations, which can decrease the steps of provenance tracking based on the elements of data provenance and support the descriptive power of the language for complex provenance tracking. Experimental results show that complex provenance tracking by the language is efficient and ease to use.