The Ministry of Social Affairs & Employment in the Netherlands conducts around 30 research studies on labor agreements per annum, with the topics varying from flexible labor market, diversity in the workplace, and the impact of the Covid-19 crisis on the collective labor agreement. The aim of these studies is twofold: on the one hand the collective labor agreements are a gauge for policy, and at the same time they can form an inspiration. The current study method is labor-intensive, takes around 4-5 months to complete and does not cover all labor agreements in the Netherlands (100 largest collective labor agreements, with a coverage ratio of 85%). To automate and enhance the process, DEUS, together with the experts from the ministry, has built Salus. Capitalising on the recent advances in natural language processing, all labour agreements have been made searchable and question answering (QA) models have been built to process research questionnaires in Dutch. This has led to a significant increase in completeness of labor agreement processing. Furthermore, agreements can be processed at once, allowing for faster and more complete research and freeing up time for the experts to focus on the in-depth data analysis.
PhD
DEUS
https://deus.ai/
Sophia Katrenko, Head of Data Science at DEUS, has worked in advertising, research, finance, publishing, and continues exploring Data Science & Analytics in other sectors, preferably with a measurable impact on society. Sophia is particularly interested in bridging the gap between research and industry, and promoting knowledge valorisation.
Dutch Ministry of Social Affairs and Employment
https://www.government.nl/ministries/ministry-of-social-affairs-and-employment
