Unilever-supported team becomes prize winner in WUR Hackathon

In this two-day pressure cooker challenge, plant scientist met front end developers, animal scientists met machine learning experts and students met agri-business partners, to collaboratively assess data, build prototypes and code API’s. All of which aimed at addressing pressing issues in life sciences.

One of the five cases for the Hackathon was put forward by Unilever Food Research. A big challenge for this department is the production of foods with a reduced environmental impact, while still delivering a high quality and great flavours. Non-animal originated and appreciable flavours often make use of an assembly of specific chemical reactions, collectively named as the Maillard reaction.

Although there are countless publications on experiments with the Maillard reaction in different conditions, it remains an unpredictable process due to its complex character. However, if efficiently processed, all this research data combined might hold clues on how to model the reaction. And that’s where the hacking and data wrangling comes in.

Unilever asked the participants: “Can we create a model that predicts if an article or paragraph is relevant when looking for a specific outcome of the Maillard reaction? Can we automatically create an overview of all experimental conditions under which the Maillard reaction was tested and link these conditions to outcomes?”

‘UniSearch’, a team of 6 students and 1 freelancer, took on this case, guided by Unilever employees. Under intense time pressure the team was expected to understand the challenge, deep-dive data, seek a data driven solution and prototype it. “The group was fuelled with strong enthusiasm, lots of energy and, fortunately for me, some good chemical background. That helped to check the relevancy and keywords”, says Jan Koek, Flavour Chemistry Scientist at Unilever.

After 32 (almost sleepless) hours and meals behind the computer, the team pitched their solution in front of a jury. They proposed to include full text in their search engine and reinforce learning by building an interactive search engine. Their creativity, technical skills and fervent commitment impressed everyone that had the opportunity to see them work. They were awarded second place and €3000,-.

For Unilever it was great to engage with students and see how they dedicated themselves zealously to the project. The hackathon vibe was full of energy, enthusiasm and synergy, despite the diversity of the groups. UniSearch even has the intent to continue with the project. Some of the students will follow the course on Big Data at Wageningen University and plan to work on our challenge as a case study during the course. We might even see them grow into a start-up with funding and guidance from Unilever Venture Funds.