The goal of Data Science projects is to extract knowledge and insights from collected data. The focus is put on the novelty and usability of the obtained insights. However, the impact of a project can be seriously reduced if the results are not communicated well. In this paper, we describe a means of managing and describing the outcomes of the Data Science projects in such a way that they optimally convey the insights gained. We focus on the main artifact of the non-verbal communication, namely project structure. In particular, we surveyed three sources of information on how to structure projects: common management methodologies, community best practices, and data sharing platforms. The survey resulted in a list of recommendations on how to build the project artifacts to make them clear, intuitive, and logical. We also provide hints on tools that can be helpful for managing such structures in an efficient manner. The paper is intended to motivate and support an informed decision on how to structure a Data Science project to facilitate better communication of the outcomes.
CITATION STYLE
Rybicki, J. (2019). Best practices in structuring data science projects. In Advances in Intelligent Systems and Computing (Vol. 854, pp. 348–357). Springer Verlag. https://doi.org/10.1007/978-3-319-99993-7_31
Mendeley helps you to discover research relevant for your work.