Topic modeling for a survey answer dataset. Topics were created using the Non-negative Matrix Factorization algorithm.
Dataset was extracted from a survey of the EU council regarding their website. Questions were the following:
- Do you have any other comments or suggestions for how we could improve the Council website?
- What changes would you like to see in order to come back?
- In what way would machine translation impact the way you use the Council website?