Abstract
Research data is very valuable, and its collection is time-consuming, work-intensive, and expensive. Therefore, it has become good scientific practice to make as much data as possible publicly available, enriched with metadata and additional information. However, the metadata of data science processes can rarely be determined automatically. Furthermore, there is often no link between a specific ...
Abstract
Research data is very valuable, and its collection is time-consuming, work-intensive, and expensive. Therefore, it has become good scientific practice to make as much data as possible publicly available, enriched with metadata and additional information. However, the metadata of data science processes can rarely be determined automatically. Furthermore, there is often no link between a specific query/research result and the associated (provenance) data and metadata. In this article, we address this challenge in data-driven science applications with a focus on marine data processing. An important goal is to trace all relevant information from a concrete research result back to its original data and meta-data. Therefore, we aim to integrate and make accessible the workflow/data provenance throughout the entire process. The main contributions of this paper are twofold: (i) we summarize the challenges in the field of interdisciplinary research projects, and (ii) we present a first approach that combines both provenance types in one concept.
More literature (via CORE)