OPENSKIMR - Open European
Skill Match Maker
is a project funded by the European Union
that aims to bring together talents,
jobs and learnings to support people
in creating their personal career routes.

“Sometimes it is a clash between real world problems and mathematics”

3 Questions to Andreas Kofler*, mathematician of OPENSKIMR

What does OPENSKIMR mean to you?


From a mathematician’s point of view, it is a combination of real world problems, like career choices, and the attempt to solve them with the help of mathematics. You might even say it is a clash between those parts (laughs).


Which steps are involved in the development of the algorithm?


At the beginning there is the problem of understanding of what goal the actually is, and how to formulate it as a mathematical problem. Then, once the mathematical framework has been chosen, usually you look for proper existing algorithms and you try to apply, adapt or extend them. In some cases, you need to create new ones. However, the first step is to look for already existing solutions to a specific problem.


Afterwards you need to test the algorithm. We still work on the base of theoretical assumptions because we have not yet collected enough real data. We try to keep the assumptions as general as possible and test the behavior of the algorithm concerning computational time and the plausibility of the results.


We have designed different algorithms which we are going to test. For example, the route planer algorithm has various functions integrated, like the matching function between user and jobs, a learning function that changes the profile once the user has successfully attended training as well as a graph construction routine which builds up different learning paths to an updated profile. We also work with clustering algorithms to organize the huge amount of skills, occupation, jobs, users or learnings.


However, the biggest challenge is to understand what the visionaries behind the project want discussed and suggest possible solutions. Also, it is important to keep an overview of how different components within the project are connected to each other.


What are the biggest challenges regarding the matching algorithm? And how do you solve them?


A main challenge is to make the right assumptions. We plan to aggregate a lot of data in the whole of Europe and we don’t know how the algorithm behaves once filled with such large amounts of data.

This is why we formulate our assumptions as general as possible in order to be as flexible as possible to adapt the algorithm.


*Andreas studied technical mathematics at University of Innsbruck and worked in several research projects. At OPENSKIMR he is responsible for the design of the matching and the route planner algorithm.


Contact us