Machine learning is a branch of artificial intelligence based on the idea that systems can learn from data, identify patterns and make decisions with minimal human intervention. RosterBuster is now building on this technology to process airline rosters faster and more accurately.
Because of new computing technologies, machine learning today is not like machine learning of the past. It was born from pattern recognition and the theory that computers can learn without being programmed to perform specific tasks. The iterative aspect of machine learning is important, because as models are exposed to new data they are able to independently adapt. They learn from previous computations to produce reliable, repeatable decisions and results.
Proof of Concept
The ability to automatically apply complex mathematical calculations to big data is a recent development. Our roster experts recognized this as an opportunity we could leverage from for the interpretation of airline rosters and specificly the understanding the hunderds of different duty codes airlines have. These are full of cryptic codes that all have very specific meanings.
Hugo Vink, who is a student of The Hague University, did the research and works on the implementation of the Proof of Concept (POC) on the AWS platform. Hugo explains: “We have a lot of historic data we can feed into our model. That allows us to predict duty code meanings with great accuracy. We then combine this with our existing logic to parse and interpret rosters.”
This will make it a very powerful approach to process airline rosters, which in turn means that users will have a much better and richer experience when they browse through their events in the app. The first benefits of his efforts are expected to be applied as early as next month.
Hugo’s contribution is another great example of the innovation that we achieve within the ecosystem of the Dutch Innovation Factory.
Image via www.vpnsrus.com