Using Azure AI Document Intelligence to Accelerate Data Ingestion and Extraction

[@shikhabasak4852]

Hi, I have a doubt with the result confidence part. In both extraction and classification models, how does azure set the confidence level? When did the user confirm (validate) that the information the model read and understood was correct or not, based on which the model is x% confident about the result? (I'm thinking of it in terms of y_pred and y_actual we do during model training on python using sklearn - which helps the model evaluate accuracy, recall, etc)