Setup Dataset

Import Data

Once we finish the essential configuration of SmartAnno, we will be asked to import data–usually txt files.

Follow the file navigator to locate the directory where the input documents are stored, and click “Confirm”.

Also, we can specify how many documents you want to import (instead of importing all of them in that directory).

Import annimation

Name your dataset

After importing, we can give a name for this imported dataset (document-level) Set Dataset Name

The SmartAnno will also ask if we want to split sentences to create a sentence-level dataset. We have two options: using TextBlob sentence splitter or PyRuSh. The PyRuSH is a sentence segmenter customized for clinical text.

Select an dataset

If you chose the sentence splitting above, you will see two datasets:

Document-level dataset: you will review and annotate per document.
Sentence-level dataset(with a suffix ‘_sents’): you will review and annotate per sentence.

Select Dataset