Reproduce “Knowing when to stop: insights from ecology for building catalogues, collections, and corpora” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jan Hajič jr., Fabian Moss
Hajič jr. and Moss apply 'Unseen Species' models from ecology to digital humanities collections, specifically the Cantus database of Gregorian chant, to estimate the proportion of unobserved repertoire and analyze the rate of new discoveries over time. Their analysis reveals that 20-50% of the repertoire remains unobserved and that the rate of new chant discovery has remained constant at about 5% over decades of cataloguing.
Key insights
- A structural analogy exists between the 'unseen data' problem in digital humanities collections and the 'unseen species' problem in ecology, allowing for cross-disciplinary methodological transfer.
- For the Cantus project, the point of diminishing returns for discovering new Gregorian chant repertoire has not yet been reached, with new chant types being found at a consistent rate over decades.
- Computational models can provide empirical estimates for collection completeness and inform strategic resource allocation and data collection in cultural heritage projects.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.