Extract alphanumeric entity

Rasa Open Source
1

Hi all,

Hope everyone doing good. I am looking how i can extract alphanumeric entity. EX: Mine customer id is [abc1199191991]

Note: Mine customer id have 19 digit

I tried to use regex but no luck it’s not extracting. If anyone have how it’s setup let me know. I tried below pipeline

language: “en”

pipeline:

this is using the spacy sklearn pipeline, adding duckling

all components will use their default values

  • name: “tokenizer_whitespace”
  • name: “intent_entity_featurizer_regex”
  • name: “ner_crf”
  • name: “ner_synonyms”
  • name: “intent_featurizer_count_vectors”
  • name: intent_classifier_tensorflow_embedding intent_tokenization_flag: true

NLU DATA: “regex_features”: [ { “name”: “Coustomer_ID”, “pattern”: “^[a-zA-Z0-9]+$” }]

Thanks

2

You have to add some examples of this entity in your NLU training data too

3

Hi Akela,

Thanks for your response. Can i use regex and duckling together. I have a customer id as numeric and id as alphanumeric.

4

Yes you can use both