0
votes

I'm working on programmatically building a LUIS model from the Microsoft Cognitive Services APIs (http://www.luis.ai/) and would like to be able to upload a file and identify the intents/entities. Is this what an entity extractor does?

I tried out the Console for the Programmatic APIs and still am not sure what I should be looking at. Found here: https://dev.projectoxford.ai/docs/services/56d95961e597ed0f04b76e58/operations/56f8a55119845511c81de46a/console

Any advice?

1

1 Answers

1
votes

Actually you need to:

  1. Build a LUIS app.
  2. Add the entity extractors you want.
  3. Add some examples and label them (mark entities in each example).
  4. Train your model; you can do so either using the LUIS website, or programmaticaly (I recommend using the website for better understanding of the model building process).

After you have a ready model, you can publish it and obtain an endpoint URL, so you can hit that URL with new examples to see how LUIS identifies the intent of the example and what are the extracted entities.

More info and examples here