|Training of Natural Language Query in Discovery: Every sample question for training can be added only once. In a real world scenario with automated feedback, similar questions / training data will be added more than one time only. Right now the training gives back error messages if a already known query is added again. It must be possible to add them again and again in order to improve performance and to react on changes of the data over time.|
Why is it useful?
|Who would benefit from this IDEA?||Customer|
How should it work?