Table of Contents
Open Table of Contents
TL; DR
Between August 2020 and August 2021, I engaged in a research project alongside my employment at Allplan Software Engineering GmbH. The project, which required a commitment of 20 hours per week, was conducted in partnership with the PEEC research group at the University of Applied Sciences Upper Austria. The focus of the project was to investigate the automated analysis of collaboration processes, and I was responsible for the implementation of the prototype.
During the project, we utilized various AI services and frameworks, with the implementation of the prototype being mainly done through programming languages C# and Python. The project’s successful outcome was a research paper that was published in 2022. It is noteworthy that the research was carried out in collaboration with Microsoft Research Cambridge.
How I came to the project?
During my Master’s thesis at the University of Applied Sciences Upper Austria, I came across the job advertisement seeking software developers for this research project. The position was particularly timely, given the changes in collaboration due to home office and hybrid work models.
Despite being employed at Allplan Software Engineering GmbH, I negotiated a reduction in my working hours to take on the research project. Following the completion of my Master’s degree in July, I began working on the project in August.
The challenges
The automated analysis of collaboration processes may seem simple at first glance, but it is a very complex task.
The aim is to recognize and analyze the different types of collaboration processes in image and video material. However, this also turns out to be a challenge for humans, which is why the analysis so far has been a time-consuming task. However, this is also a challenge for automated systems as the different processes are sometimes difficult to distinguish from each other. This task is further complicated by the way collaboration has changed, especially in recent years through hybrid or remote collaboration.
The project
Throughout the course of our project, our team worked in close collaboration with researchers at Microsoft Research Cambridge. We engaged in a dynamic and ongoing exchange of ideas and strategies aimed at finding innovative solutions to our project’s objectives.
My main role in this project was that of a software developer, responsible for the creation and implementation of the system. The implementation of the system was undoubtedly a crucial aspect of the project, but equally important was the investigation into the potential for automatic recognition of various collaboration processes.
As part of our work, we had the opportunity to experiment with diverse technologies and cutting-edge AI services, allowing us to compare and contrast their effectiveness in relation to our project’s goals. Additionally, we were able to use partially experimental frameworks developed by Microsoft in our prototype.
By the end of the year we successfully completed the prototype. To test its functionality, we conducted a comprehensive user study in which the prototype was thoroughly assessed. Ultimately, our work resulted in a research paper, published in December of 2022, which captured our project’s contributions and achievements.
Paper
T. Neumayr, M. Augstein, J. Schönböck, S. Rintel, H. Leeb and T. Teichmeister, “Semi-Automated Analysis of Collaborative Interaction-Are We There Yet?”, Proceedings of the ACM on Human-Computer Interaction (PACM HCI), vol. 6, no. ISS, pp. 354-380, 2022, doi: https://doi.org/10.1145/3567724.