0
$\begingroup$

Excuse if this has been answered before.

I need to extract features and parse from a piece of text and run some analysis. For e.g. "Plot the past 5-year sales of Apple" should give me the following

Information:

  • Company: Apple
  • Item: Sales
  • Period: past 5 years
  • Action: Plot

What deep learning techniques / algorithms should I be looking to use?

Any pointers are highly appreciated.

$\endgroup$
1
$\begingroup$

This is usually handled as a two step process:

  • intent detection
  • slot filling

Intent detection is basically just a kind of classification. So if your program has multiple kinds of questions it can be asked you build a list of examples for each and train a classifier.

Slot filling is typically modelled as a sequence labelling problem, so you can look into seq2seq.

You can find more detailed information in any tutorial on how to make a chatbot.

$\endgroup$
  • $\begingroup$ Thanks polm23. I dont have a dataset for this task. Any inputs on how to create one? $\endgroup$ – user6027414 Jun 26 '18 at 9:50
  • $\begingroup$ Pretend you are a user of your app and make some fake data. That should be enough to get you started. $\endgroup$ – polm23 Jun 26 '18 at 10:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.