23  Reproducible research

Aviation like many other scientific domains suffers for lack of data openness and lack of reproducibility of published research [1]. Many reasons are made up not to disclose data, the main ones being sensitivity and ownership.

24 Produce meaningful maps

Tip

Author: Enrico, ?

Note

Author’s note: The most overlooked aspect of data analysis probably turns around data sharing. Data curation is often a very time consuming process and enriching data by labelling specific tags or merging several sources of information brings additional value to a dataset. This part deals with the data sharing and publication process. (paper reproducibility?)

Introduction about reproducible research. This would also include sharing the code beyond the dataset cleaning to producing related summaries, tables/graphics, etc.

1. Good practices for producing a clean dataset > #include and use some of the datasets in ‘traffic’ / Zenodo / > … these could be used in the examples in the book (and it would > be great to explain how to arrive to a clean dataset)

2. Issues around dataset sharing > open data, licensing, etc.?

[1]

M. Bourgois and M. Sfyroeras, “Open Data for Air Transport Research: Dream or Reality?” in Proceedings of The International Symposium on Open Collaboration - OpenSym ’14, 2014, pp. 1–7, doi: 10.1145/2641580.2641602.