I am a maker of digital products currently living in Amsterdam.
As a data scientist with originally an interaction design background,
I enjoy working with machine learning data problems that require human input.

How realtime vehicle classification can improve the travel experience of public transport

The current user experience of public transport planning is not optimized for use on-the-go. Common solutions are limited to planning from a static location, such as train stations or addresses. This approach assumes that passengers plan their journey beforehand and does not take into account possible disruptions. The planned route is so to speak "static".
To support passengers on their planned journey during disruptions, I created a rough classification algorithm that attempts to find the vehicle the passenger is aboard of. The input for classification is the current passenger coordinates and the current coordinates of all active public transport vehicles in The Netherlands. As soon as we know with which vehicle the passenger is traveling, we can check if the vehicle's route is disrupted and what consequence this has for the passenger.

Start the animation below to see how classification can be done.
In addition to updating a travelers journey in case of disruptions, there are many other things we can do with this functionality. For example, an automated message can be sent as soon as the traveler experiences a disruption. A full overview of the possible use cases I could think of so far is listed below.
Share your trip
Share trips with friends to let them know which train you're on and if you have a delay.
Order food on the go
Do not feel like FASTfood? Avoid waiting times? Order food from the train and pick it up once you're at the station.
Receive reminders
Need to pickup a package at a train station? Notifications can be send when you're near it.
Plan routes on the go
Plan routes with the current vehicle as starting point rather than calculate when you arrive somewhere.
Pro-active functionality
Every Monday at 8' on the same train? Maybe you're heading towards the same destination again.
Send messages in case of disruptions
Automatically update your colleagues when you will be late for a meeting because of a disruption.