From traditional methods to AI: Researchers gain new insights
The development of artificial intelligence (AI) is advancing rapidly, and it is affecting research. The use of AI and machine learning is not just another option; it is a necessity for research in the digitalized society. This was the message conveyed during the workshop "Research using NLP and Text Analysis" at the School of Business, Economics and Law at the University of Gothenburg, organized by the research cluster AI and algorithm driven research.
The workshop aimed to inspire new ideas and opportunities with AI, while also providing participating researchers with a better understanding of the boundary between quantitative and qualitative research. With the proliferation of digitalization, there is so much data that traditional research methods are insufficient. Hence, AI plays a crucial role in research. At the same time, the forum was intended to be a source of inspiration where participants could be inspired by one another.
– By seeing other methods, we can probably learn more and get ideas about how we can further develop our research. We should avoid reinventing the wheel over and over again, says Mari Paananen, Associate Professor of Accounting at the School of Business, Economics and Law.
Tools for researchers to overcome limitations
To bridge the gap between qualitative and quantitative research, Computation Methods (CM) were presented as a tool that also offers opportunities for research to study human behavior in new ways.
Fredrik Svahn, Associate Professor at the Department of Applied Information Technology, talked about the possibilities that Computation Methods offers researchers to overcome limitations with quantitative and qualitative methods.
– An important point of this workshop was that I invited researchers who use qualitative methods, so that they can see the boundary between quantitative and qualitative research and actually utilize data, says Mari Paananen.
Using AI technology to identify political dog whistles.
Asad Basheer Sayeed, an associate professor of computational linguistics at the Department of Philosophy, Lingqistics and Theory of Science, is working to identify "political dog whistles" and leveraging technology to do so efficiently. Political dog whistles refer to coeded or persuasive language in political communication amied at gaining support from a specific group without provoking opposition.
Asad Basheer Sayeed presented the project GRIPES, which analyzes the boundaries of language manipulation, such as political dog whistles.