PACMan2: Next Steps in Proposal Review Management

Abstract

With the start of a new Great Observatories era, there is renewed concern that the demand for these forefront facilities, through proposal pressure, will exceed conventional peer-review management's capacity for ensuring an unbiased and efficient selection. There is need for new methods, strategies, and tools to facilitate those reviews. Here, we describe PACMan2, an updated tool for proposal review management that utilizes machine learning models and techniques to topically categorize proposals and reviewers, to match proposals to reviewers, and to facilitate proposal assignments, mitigating some conflicts of interest. We find that the classifier has cross-validation accuracy of $80.0\pm 2.2\%$ on proposals for time on the Hubble Space Telescope and the James Webb Space Telescope.

arXiv comment10 pages, 7 figures

References in corpus (8)

• Scikit-learn: Machine Learning in Python
• Astropy: A Community Python Package for Astronomy
• Patterns of Text Reuse in a Scientific Corpus
• Distributed peer review enhanced with natural language processing and machine learning
• Knowledge discovery through text-based similarity searches for astronomy literature
• The Unified Astronomy Thesaurus: Semantic Metadata for Astronomy and Astrophysics
• Peer-review under review - A statistical study on proposal ranking at ESO. Part I: the pre-meeting phase
• The Proposal Auto-Categorizer and Manager for Time Allocation Review at Space Telescope Science Institute