Natural language processing (NLP) is used for tasks such as sentiment analysis, topic detection, language detection, key phrase extraction, and document categorization.
Sentiment Analysis is the process of determining whether a piece of writing is positive, negative or neutral.
Reference:
https://docs.microsoft.com/en-us/azure/architecture/data-guide/technology-choices/natural-language-processing

Reference:
https://docs.microsoft.com/en-us/azure/cognitive-services/qnamaker/concepts/data-sources-and-content

1 - data preparation
2 - model training
3 - model evaluation
Reference:
https://docs.microsoft.com/en-us/azure/machine-learning/concept-ml-pipelines

1 - Ingest and prepare a dataset.
2 - Split the data rendomly in training.. .
3 - Train the model.
4 - Evaluate the model against the validation dataset.