Open glacierck opened 2 years ago
Thank you for using MedTator!
The annotation schema design is always challenging, and the attribute design is also difficult.
The first idea may be to use a good editor and enable the word wrap to remove the horizontal scroll bar. :p Then, the annotation guideline design may also be updated to reduce the burden on annotators. I guess it will be very difficult for annotators to select one value exactly from a huge list, and it may result in annotation errors easily. So, the huge attribute can be split into several sub-attributes.
For example, the country
can be split into Asia
, Americas
, Africa
, etc.
In addition, we are also considering using new format for the schema file, such as YAML or JSON since they are easier to edit and use.
I have an attribute about "country", which has about 200 enumeration values. This will make the DTD difficult to read. Is there any good solution to this kind of attribute type.