Towards an AI-assisted world Flora? Some personal thoughts.
- Gather a large dataset of plant images, taxonomic descriptions, geographic distributions, etc. This could come from existing sources like herbarium records, taxonomic literature, citizen science observations, etc. AI can help process and standardize this data.
- Use AI models like computer vision and natural language processing to automatically extract visual features, taxonomic characters, and locality information from the dataset. This allows rapid parsing of large amounts of data.
- Have the AI models automatically generate draft species descriptions, keys, distribution maps and other content for the world Flora using the extracted data. This provides a baseline to build on.
- Expert taxonomists and editors then review the AI-generated content, correct errors, fill in gaps, and finalize the descriptions, keys and images. The goal is to produce a high-quality world Flora.
- Set up a system for continuous updating as new data comes in. AI models can suggest changes and additions for human review. This keeps the Flora dynamic.
Is this feasible?
