ARABIAN PENINSULARed SeaGulf of AdenArabian SeaGulf of OmanThe GulfStrait of HormuzBab el-MandebAREA OF STUDYSimplified outline · approximate

Project report · In progress

Maritime choke points

How news, social media and shipping data around maritime choke points relate.

StatusIn progress · exploratory
RegionsHormuz, Red Sea, Gulf, Arabian Sea
SourcesGDELT, Bluesky, AIS, UKMTO
Built so farFilter (first version), sampler

Currently building. Exploratory, and the direction may change.

On this page

Overview

I’m studying how news, social media and shipping data around the world’s maritime choke points relate to each other: the Strait of Hormuz, the Red Sea, the Arabian Peninsula / Gulf and the Arabian Sea.

Data sources

  • GDELT GKG: global news, filtered by location around these areas and by market topic
  • Bluesky: social posts, filtered by keywords
  • AIS feeds: how ships are actually moving
  • UKMTO alerts: maritime security incident reports

The idea

Watch these streams, detect spikes in activity, and when a spike happens, read what the news is saying and compare it with what ships and maritime alerts are doing at the same time.

Honestly, we don’t yet know how these signals relate, or where the project will end up. We’re building it because we’re curious, and building it is how we find out.

Planned system

A high-level sketch of where this is heading. The filter (first version) and the sampler are built; everything else is planned.

  1. Data sourcesPlannedGDELT GKG news, Bluesky posts, AIS ship positions, UKMTO alerts
  2. FilteringBuiltGDELT news by place and by market topic, first version. Filtering Bluesky posts by keywords is still planned
  3. Spike detectionPlannedNotice when activity jumps
  4. SamplerBuiltTurns thousands of headlines into a short, ranked report of topics
  5. Study alongside AIS + UKMTOPlannedCompare what the news says with what ships and maritime alerts are doing

Status

In progress and exploratory. There are no results to report yet.

The first real run of the sampler, on 15 November, is written up as a news report.

The code isn’t public yet. A GitHub repo will follow once it’s been cleaned up and properly debugged.