Detect entities that are not included in training data

Rasa Open Source
1

I am building a chatbot that searches a content library using an entity “search_string” extracted from user input. However, the chatbot only seems to be able to extract entities if they are included in the training data. I do not want to use a lookup table, since “search_string” should be able to be anything.

My hope is that the chatbot will be able to detect “search_string” regardless of whether it is included in the training data.

My nlu.yml looks like:

- intent: search_content
  examples: |
    - i want to search content for [maths](search_string)
    - search content for [english](search_string)
    - content for [lesson planning](search_string)
    - find me content for [classroom management][search_string] online
    - get content results for [equality](search_string)
    - search for [geography](search_string) content
    - i need content for [science](search_string)
    - i want [wellbeing](search_string) content
    - get [assessment](search_string) content
    - get content for [languages](search_string)
    - find [behaviour management](search_string) content
    - [digital literacy](search_string) content
    - content for [english and maths]
    - search content for [communication](search_string)
    - content for [planning](search_string)
    - find content for [inclusion](search_string)
    - i need [remote and online learning](search_string) content
    - i want [blended learning](search_string) content
    - get content for [maths](search_string)
    - i need [flipped learning](search_string)[content] content
    - do you have any [english](search_string) content

I also have a “search_string” slot defined in domain.yml:

search_string:
  type: text
  influence_conversation: true
  auto_fill: true

Finally, my config.yml file is unchanged from default and looks like:

pipeline:
# # No configuration for the NLU pipeline was provided. The following default pipeline was used to train your model.
# # If you'd like to customize it, uncomment and adjust the pipeline.
# # See https://rasa.com/docs/rasa/tuning-your-model for more information.
#   - name: WhitespaceTokenizer
#   - name: RegexFeaturizer
#   - name: LexicalSyntacticFeaturizer
#   - name: CountVectorsFeaturizer
#   - name: CountVectorsFeaturizer
#     analyzer: char_wb
#     min_ngram: 1
#     max_ngram: 4
#   - name: DIETClassifier
#     epochs: 100
#     constrain_similarities: true
#   - name: EntitySynonymMapper
#   - name: ResponseSelector
#     epochs: 100
#     constrain_similarities: true
#   - name: FallbackClassifier
#     threshold: 0.3
#     ambiguity_threshold: 0.1

# Configuration for Rasa Core.
# https://rasa.com/docs/rasa/core/policies/
policies:
# # No configuration for policies was provided. The following default policies were used to train your model.
# # If you'd like to customize them, uncomment and adjust the policies.
# # See https://rasa.com/docs/rasa/policies for more information.
#   - name: MemoizationPolicy
#   - name: RulePolicy
#   - name: UnexpecTEDIntentPolicy
#     max_history: 5
#     epochs: 100
#   - name: TEDPolicy
#     max_history: 5
#     epochs: 100
#     constrain_similarities: true

Thanks for any help on this!