Provenance Inference Techniques: Taxonomy, comparative analysis and design challenges

Provenance has many applications in assessing the data quality, computational efficiency, security, and storage reliability. Provenance inference (PI) is the process of forming conclusions derived through any evidence or reasoning by static code analysis. However, despite its manifold applications, the subject of PI has not been thoroughly studied and emphasized upon. The main objective of this article is to provide a comprehensive review of the available literature on provenance inference techniques (PITs). To achieve this, we first identify the needs and requirements essential for PITs. Then, a thematic classification of the existing PITs is proposed in form of taxonomy. Moreover, we perform a comprehensive comparative analysis by highlighting the strengths and weaknesses of the existing literature on PITs. Furthermore, we have identified a set of design challenges, which should be taken into consideration. Finally, we conclude the paper by presenting recommendations, issues and open research challenges that should be considered by future research studies in this domain.