Presented By O'Reilly and Cloudera
Make Data Work
31 May–1 June 2016: Training
1 June–3 June 2016: Conference
London, UK

90% of the world's trade is transported by sea, but what data do we have about ship activity worldwide?

Tal Guttman (Windward)
11:15–11:55 Friday, 3/06/2016
Data innovations
Location: Capital Suite 4 Level: Non-technical
Average rating: *****
(5.00, 1 rating)

With over 90% of the world’s trade transported over the oceans, data on ship activity is critical to decision makers across industries. But despite the huge stakes at sea, ship activity remains a mystery: the data is massive, fragmented, and extremely unreliable when taken as is.

Anyone trying to make sense of maritime data faces three key challenges:

  1. Data is dependent on human inputs, be it a ship’s crew transmitting its location or an agent at a port recording ship arrivals and departures, making the data vulnerable to both human error and outright manipulation.
  2. Data is corrupt. The existence of commercial satellites that pick up ship transmissions was a big factor in the creation of maritime big data. However, the process of picking up these transmissions and then relaying them back to base stations and to end consumers results in significant data corruption upwards of 50%.
  3. Context is key. Even after you’ve solved the problems of data manipulation and corrupt data, how do you assess the meaning of a given data point? For example, if a ship is in a certain location, is that consistent with its historic behavior or a suspicious deviation from its pattern of life? Does the vessel’s route make economic sense or does it raise suspicion because it violates shipping economics principles? To answer these questions, you need to bring a new, data sciences perspective to maritime data. It is critically important to fuse data from multiple sources, flag data manipulation, and organize the data into unique DNAs—the full life history of each vessel worldwide—to provide the context to understand what is happening at sea.

Tal Guttman explores how data science can shed light on this critically important but opaque world.

Photo of Tal Guttman

Tal Guttman


Tal Guttman is director of the Core Platform at Windward, a maritime data and analytics company. In this role, Tal leads a team building the world’s first maritime data platform, which is analyzing and organizing the world’s maritime data and making it accessible and actionable across the ecosystem, from flagging criminal threats at sea to combatting illegal fishing to identifying new market trading opportunities. Prior to joining Windward, Tal was head of strategic innovation for the IDF’s Elite Intelligence Corps unit, 8200. Tal holds a bachelor’s degree in history from Tel Aviv University. He lives in Tel Aviv with his wife and two sons.