What's new in v2.0.0:
Classification prediction with the Maelstrom Research (MR) Taxonomy
Variable classification based on the BioBert model, a biomedical language representation model designed for biomedical text mining tasks is available. The BioBert model was fine-tuned with classified variables of epidemiological studies from the Maelstrom Research catalogue.
Automated classification
The variables of a data set can be classified automatically with predictions from the Maelstrom Research Taxonomy.
Search
- The search is extended by an autocomplete function
- Filter for only allowing annotations with MR tags (not domains)
Import/Export
- Import and export of CSV files is available
- The imported file is checked for the OPAL format and the user is informed about missing columns
- The export of annotations in the OPAL format is avialable
Multiple variable annotation
Annotation of multiple variables at once is available.