0
$\begingroup$

I've been working on text -> intent -> command execution for a particular application and while I've found many papers and code that work well for intent classification (1, 2, etc.), they stop there. For example, given a standard music intent "Play some music by U2" such classifiers return me an intent class play_music. But the information about the artist, track requested aren't part of it. Everytime I search for papers on text -> intent, I end up with classifiers instead of any deep learning models that can not only classify, but maybe give me an importance vector to extract the key parameters from my intents to pass onto the intent class.

Does anyone know of any papers or implementations like this?

Note: I'm not looking to make a voice assistant but a text->command execution engine for a very specific purpose.

Improve this question
$\endgroup$
2
  • $\begingroup$ Can you please put your specific question in the title? $\endgroup$
    – nbro
    Apr 13 at 20:47
  • 1
    $\begingroup$ @nbro changed it to better reflect the intent. $\endgroup$
    – Ani
    Apr 18 at 17:34

1 Answer 1

Reset to default
0
$\begingroup$

Found it! After a lot of searching and reading through different papers, this was the exact one I needed.

"BERT for Joint Intent Classification and Slot Filling"

There's also an example dataset based on Snips that shows how to create custom intents, entities, slots and train different kinds of transformers to classify text and extract words to fill the appropriate slots.

Improve this answer
$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .