Haaga-Helia’s students are developing a new kind of recommendation library for Yle Areena. The recommendation library help users discover content that they may not come across otherwise.
Haaga-Helia’s students are developing a new kind of recommendation library for Yle Areena. The recommendation library recommends content based on user’s likes and helps them discover new content they might not come across otherwise.
The idea for the project came from Teemu Havulinna and Ohto Rainio, who teach computing science at Haaga-Helia. They wanted to offer students a meaningful and work-oriented course project that would increase students’ work experience and provide them with work-life contacts already during the studies.
Yle’s Development Manager Jouni Frilander says that it was easy to accept the proposal coming from the teachers. He already had previous experience of a successful collaboration with Haaga-Helia.
Although Yle does regular collaboration with Finnish universities, Frilander thinks there could be even more. Working with students is, in his opinion, not only fun but also instructive.
- Truth to be told, we also get fresh ideas from these cooperation projects, he says.
According to Frilander, Haaga-Helia’s students solve challenges independently and their enthusiasm is contagious. However, teachers also play an important role in the success of the collaboration.
The professional guidance of Haaga-Helia's teachers has had a positive effect on the whole experience of working together. He also thinks that the collaboration will increase the interest in Yle as a future intership and workplace.
- We want to show students that we have interesting challenges and tasks to offer, he says.
As a development manager, Frilander wants to develop Yle Areena into an even more meaningful service that would reach every Finn, regardless of age or gender.
The body of the recommendation algorithm, developed by Haaga-Helia's students for the software project, is written by lecturer Ohto Rainio. It all started last fall when he took part in an Azure pilot course organized by Erasmus+ -funded IoT Rapid-Proto Labs and run by Houston Inc.
The people selected for the pilot were able to take advantage of the consulting company’s Azure Machine Learning Studio software. The simplicity of it surprised Rainio.
- I was seriously surprised at how easily such a recommendation algorithm could be done, he says.
The ease got him excited and led him to become more familiar with the topic. And that eventually led him to learn a completely new programming language as well as to the creation of his algorithm utilized in student projects.
According to Rainio, the algorithm arose from need.
Together with his colleague Teemu Havulinna, he teaches an open-source software development course, where students can utilize the codes in the open distribution in work-based projects.
- I wanted to use my code to encourage students to develop and share their codes as well, he says.
