PROJECT GOAL

Making 658+ social labour contracts more accessible with natural language processing

THE CHALLENGE

The Ministry of Social Affairs and Employment is responsible for more than 900 collective labour agreements (CAOs), covering more than 6 million employees in the Netherlands. These CAOs need to be kept up to date (aligned with any changes in the law) and information needs to be retrieved from these documents on a regular basis.

The current process of analysing and information retrieval from CAO’s is largely manual and very labour intensive. The Ministry wanted to find a way to automate this process and make it more accurate, freeing up time from the Ministry specialists, enabling them to focus on more strategic tasks.

THE OUTCOME

We applied natural language processing to automate and augment the existing labour intensive process of searching and analysing collective labour agreements.

This allows staff to focus on the more rewarding aspects of the task at hand, and at the same time increasing overall accuracy.

CLIENT
Ministry of social affairs & employment
INDUSTRY
Government
TECHNOLOGIES
· natural language processing · machine learning
METHODS
· qualitative research · expert interviews · product design · agile development · pilot project · concept validation
METHOD KIT

// our process

Together with the specialists at the Ministry, we mapped out the current process of searching, analysing and processing collective labour agreements and identified opportunities to improve the process with automation.

We developed a user-friendly interface where PDF's of the 658+ collective labour agreements can be uploaded and searched.

We used natural language processing and machine learning to create a question answering module to find the most relevant information across the 900+ documents.

// the results

Together with the Ministry specialists, we redesigned and optimised the process of searching, analysing and processing collective labour agreements. 
In 12 weeks, we designed and built an end-to-end digital solution that includes search and question answering functionality.
Applying AI to the information retrieval process has resulted in an average of 25% time saved per information retrieval request.
"
The DEUS team showed us how AI can be applied in a practical way to improve the speed and accuracy of our work. We worked very closely together on the design and development of the solution which offers a great way to retrieve information from the large number of documents that our team needs to analyse

Marieke van der Putten

Senior Innovation Manager at the Ministry of the Interior and Kingdom Relations

"
FESTINA LENTE

Curious to learn more?

Reach out to us!