Paper Number
1436
Paper Type
Completed
Description
Recent advances to machine learning (ML) and its rapid proliferation spur the wide-spread development of advanced analytics applications. Nonetheless, the capabilities of (ML) can be stalled due to limited or missing data. In this regard, the production of artificial data offers a promising solution. However, its full potential is yet to be unleashed since it's frequently misunderstood or overseen. We attribute this to a lack of practical guidance on when and how to employ artificially generated data. Against this backdrop, we draw on two streams—namely, method engineering and design science to develop "GenFlow", a novel method useful to practitioners as well as researchers. The utility is demonstrated in retrospect for previous work and empirically accessed for the context of employee attrition.
Recommended Citation
Kowalczyk, Peter; Röder, Marco; Rottmann, Janine; and Thiesse, Frédéric, "Designing a Method to Nudge Analytics with Artificially Generated Data" (2023). ICIS 2023 Proceedings. 4.
https://aisel.aisnet.org/icis2023/isdesign/isdesign/4
Designing a Method to Nudge Analytics with Artificially Generated Data
Recent advances to machine learning (ML) and its rapid proliferation spur the wide-spread development of advanced analytics applications. Nonetheless, the capabilities of (ML) can be stalled due to limited or missing data. In this regard, the production of artificial data offers a promising solution. However, its full potential is yet to be unleashed since it's frequently misunderstood or overseen. We attribute this to a lack of practical guidance on when and how to employ artificially generated data. Against this backdrop, we draw on two streams—namely, method engineering and design science to develop "GenFlow", a novel method useful to practitioners as well as researchers. The utility is demonstrated in retrospect for previous work and empirically accessed for the context of employee attrition.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
11-PM