The supplementary directory contains the data and code from our survey. The Papers, Taxonomy, and Elicitation Paths spreadsheet contains three sheets. The first sheet lists the 28 papers covered in our survey. The second sheet outlines the hierarchy of the taxonomy. The third sheet contains the 73 elicitation paths coded according to the taxonomy. You can also view this spreadsheet on Google Sheets. Visualizations.tgz contains the code used to create the visualizations in this paper. You can also view this code on Observable.
@Article{kerrigan2021survey,
AUTHOR = {Kerrigan, Daniel and Hullman, Jessica and Bertini, Enrico},
TITLE = {A Survey of Domain Knowledge Elicitation in Applied Machine Learning},
JOURNAL = {Multimodal Technologies and Interaction},
VOLUME = {5},
YEAR = {2021},
NUMBER = {12},
ARTICLE-NUMBER = {73},
URL = {https://www.mdpi.com/2414-4088/5/12/73},
ISSN = {2414-4088},
ABSTRACT = {Eliciting knowledge from domain experts can play an important role throughout the machine learning process, from correctly specifying the task to evaluating model results. However, knowledge elicitation is also fraught with challenges. In this work, we consider why and how machine learning researchers elicit knowledge from experts in the model development process. We develop a taxonomy to characterize elicitation approaches according to the elicitation goal, elicitation target, elicitation process, and use of elicited knowledge. We analyze the elicitation trends observed in 28 papers with this taxonomy and identify opportunities for adding rigor to these elicitation approaches. We suggest future directions for research in elicitation for machine learning by highlighting avenues for further exploration and drawing on what we can learn from elicitation research in other fields.},
DOI = {10.3390/mti5120073}
}