This project shall contribute to the assessment of university research dissemination and knowledge transfer by developing a new automated approach that makes use of methods within the field of deep learning and in particular bag of words, RNNs and transfer learning.
It will be practical, applied and shall investigate the concrete case of DTU. The goal is to use text data from the university as well as from relevant companies to detect technology/knowledge transfer.
In the concrete case of DTU several potential data sources are available and shall be used to reveal interesting patterns (data sources are described at the end of this document).
The student is expected to be familiar with deep learning and transfer learning in particular for text data (knowledge of recurrent neural networks is an advantage).