Master Thesis – Natural Language Processing at Customer Service – Vattenfall R&D bei Vattenfall AB


The Vattenfall R&D section for Data Science & AI apply advanced analytics to support the mission of fossil free living within one generation. Key in this development is a clear focus on our customers experiences and feedback. In order to provide excellent support to our customers we are continuously looking for new ways to execute customer service.

Our skilled analysts in the Customer Service department are implementing new ways of supporting our agents to provide a good customer experience, for example forecasting the number of calls that the service center receives. Machine learning tools have an incredible potential, and next focus is Natural Language Processing (NLP). So far Customer Service has done some basic text analysis on customer feedback data, and we want to now investigate if NLP can help us to take next steps in this area.

Swedish natural language processing models need to be evaluated with a common framework, something that is being developed by AI Sweden. The student is expected to connect with this framework.


The goal of this thesis is to evaluate and deploy suitable model for Swedish natural language processing for a use case discussed together with Vattenfall. Examples of this could be: Chatbot, analyzing feedback, speech to text or directly supporting agents.

Scope of the thesis (30HP)

We would like the student to:

1. Together with Customer Service evaluate the best use case to focus on

2. Suggest and implement models together with our team working with Data Science & AI. This could e.g. be one of the below:

a. Text analysis of dialogues to identify automation opportunities

b. Text analysis of feedback to identify improvement areas

c. Speech analysis to understand automation potential for e.g., a chat bot

d. Quality monitoring, possibility to use speech analysis for coaching agents

What else do I need to know?

  • This master thesis is done as a collaboration between Vattenfall R&D and Customer Service.
  • The location for this work is flexible, the team is mostly situated at Vattenfall R&D in Solna, Stockholm and in Umeå.


You are a computer science or engineering student with experience in different machine learning models used for natural language processing. You have a keen interest in sustainability and a drive to share your knowledge by developing solutions that can be used broadly with limited analytics knowledge.

Swedish understanding is desirable since the language to be analyzed is Swedish.


CV, cover letter, and grades are to be submitted in the application.

The interview process will be ongoing, as soon as relevant applicants are identified.


Deadline for application: 2020-12-06.


Do not hesitate to contact us for any questions:

Axel Eriksson

Vattenfall R&D

Example: 9 Ways to use NLP in Customer Service

1. Accurate call routing with IVR systems

2. Routing support tickets

3. Analyzing customer feedback

4. NLP and customer service chatbots

5. Agent support with NLP

6. Business data analysis

7. Sentiment analysis and customer satisfaction

8. Speech-to-text applications

9. Built-in search bars in knowledge base

Achtung: In der Regel brauchst du die Genehmigung deiner Uni oder eines Betreuers, damit die Projekt- oder Abschlussarbeit auf Graduateland als Teil deines Studiums anerkannt wird. Bitte setz dich rechtzeitig mit den richtigen Stellen in Verbindung, damit du sicher das richtige Projekt auswählen kannst.