The dataset provides natural language data for training binary risk classification models in German language. It contains labelled text data for both risks and chances.
License: CC-BY-4.0, see https://choosealicense.com/licenses/cc-by-4.0/
- Goal: Provide natural language text data for training binary classification models in risk management applications
- Reason why: There aren't yet enough data sets publicly available that cover both German language and risk classes
- Each record of the text data set is labelled as follows: 1 = the issue is a risk, 0 = the issue is a chance
- Number of text data records labelled as "risk": 503
- Number of text data records labelled as "chance": 503