Follow a theme.
Start with a broad idea or find something specific. Each theme connects talks from across rooms and sessions.
Starting points
AI-ready data
Making biodiversity data and metadata fit for machine learning: readiness definitions and metrics, dataset documentation (e.g. Croissant, dataset cards) and FAIR-for-AI.
Specimen digitisation
Digitising natural history specimens and collections, from imaging workflows and stations to national and legacy digitisation programmes.
Persistent identifiers
Persistent and content-based identifiers for specimens, samples, media, names, literature and organisms, and their stability.
Data provenance
Recording where data came from and how they were processed, including versioning, paradata and auditability.
Knowledge graphs
Constructing and using RDF and property-graph knowledge graphs to connect biodiversity data, literature and knowledge.
Data quality assessment
Tests, flags, validation and cleaning that assess and improve the quality of biodiversity records (e.g. BDQ tests).
Standards development and governance
How TDWG and other bodies develop, review, ratify, version and sustain standards, and how standards evolve and are adopted.
Biodiversity literature mining
Extracting occurrences, traits, names and other data from biodiversity literature and grey literature with text mining and NLP.
All 123 themes.
No themes match that filter.