I\u2019m developing an open dataset that links ship-tracking signals (automatic transponder data) with registry and ownership information from Equasis and GESIS. Each record ties an IMO number to: \u2022 broadcast identity data (position, heading, speed, draught, timestamps) \u2022 registry metadata (flag, owner, operator, class society, insurance) \u2022 derived events such as port calls, anchorage dwell times, and rendezvous proximity<\/p>\n
The purpose is to make publicly available data more usable for policy analysis, compliance, and shipping-risk research \u2014 not to commercialize it.<\/p>\n
I\u2019m looking for input from data professionals on what analytical directions would yield the most meaningful insights. Examples under consideration: \u2022 detecting anomalous ownership or flag changes relative to voyage history \u2022 clustering vessels by movement similarity or recurring rendezvous \u2022 correlating inspection frequency (Equasis PSC data) with movement patterns \u2022 temporal analysis of flag-change \u201cbursts\u201d following new sanctions or insurance shifts<\/p>\n
If you\u2019ve worked on large-scale movement or registry datasets, I\u2019d love suggestions on:<\/p>\n
- \n
- \n
variables worth normalizing early (timestamps, coordinates, ownership chains, etc.)<\/p>\n<\/li>\n
- \n
methods or models that have worked well for multi-source identity correlation<\/p>\n<\/li>\n
- \n
what kinds of aggregate outputs (tables, visualizations, or APIs) make such datasets most useful to researchers<\/p>\n<\/li>\n<\/ol>\n
Happy to share schema details or sample subsets if that helps focus feedback.<\/p>\n<\/div>\n
submitted by \/u\/captain_boh <\/a>
[link]<\/a><\/span> [comments]<\/a><\/span><\/p>
