AutoPart
Semiautomated partner search support for technology transfer using machine learning methods
Prof. Dr. Peter Mandl
Department of Computer Science and Mathematics
Together with DLR, the Competence Center for Business Information Systems at Munich University of Applied Sciences is conducting research into a tool-based methodology for identifying the willingness to innovate and the need for cooperation among SMEs and medium-sized businesses in order to identify potential business partners for technology transfer in a semi-automatic way.
The scientific goal of this project is to investigate whether and how modern methods of data analysis using machine learning can facilitate matchmaking between research institutions and SMEs. Machine Learning algorithms will be used to automatically evaluate existing websites, job advertisements and company catalogues. The collected data will be used to classify potential partners according to defined criteria regarding their willingness to innovate and their need for digitisation.
The complex question of the right selection criteria will be answered by means of an application example, the technology transfer of simulation and control technology know-how. A novel crawler will be used to retrieve defined data sources at regular intervals. Selected data sources are searched for unstructured or semi-structured pages, content is extracted, stored and evaluated using machine learning methods (neural networks, random forest, bagging, boosting).
The new tool should make it possible to automate and simplify the complex, relationship-based matchmaking between research institutions and SMEs when initiating research projects.
The methodology will be designed generically so that it can be used with new selection criteria for other fields of application outside simulation and control technology. The entire process will be supported, from the preparation and configuration of the search criteria to the implementation of matchmaking and evaluation via standardised interviews. The methodology will initially be limited to SMEs, but can in principle also be applied to large industrial groups.
Running duration:
01.07.2019 - 30.06.2022
Funded by:
Federal Ministry for Education and Research
Project Executing organisation:
DLR Project Management Agency