Lookup Tables and Intent Classification

Rasa Open Source
1

I am confused by the Rasa documentation about Lookup tables and if they help Intent classification or not.

  1. Reading about Lookup tables it seems like they are only meant for Entity extraction but
  2. In the documentation for Lookup tables it also states:

When lookup tables are supplied in training data, the contents are combined into a large, case-insensitive regex pattern that looks for exact matches in the training examples. These regexes match over multiple tokens, so lettuce wrap would match get me a lettuce wrap ASAP as [0 0 0 1 1 0]. These regexes are processed identically to the regular regex patterns directly specified in the training data.

  1. Reading the Regex features documentation it states:

Currently, all intent classifiers make use of available regex features.

So do Lookup tables help with intent classification or not?

3

I have the same problem! Actually it does help in entity extraction as in when you have a list of names You can add all those names in the lookup table and the interpreter will look for the pattern or name in the lookup table! So it helps in entity extraction and I think this is inter-related to intent identification also!

4

I am looking for a definite answer (hopefully from a dev) but thanks @Akshit