Intents not as per expectation

Hi Need help n understanding how to understand why a new sentence shows a particular intent. I have trained data for six intents I got a sentence such as "yes no yes ok " I ended up getting deny as the intent.yes, ok are all in affirm training set. Why is the model giving extra weightage to the “no”. Is it loading some embeddings? I am using the following pipeline

Any help would be great…

Configuration for Rasa NLU.

language: en pipeline:

  • name: WhitespaceTokenizer
  • name: RegexFeaturizer
  • name: LexicalSyntacticFeaturizer
  • name: CountVectorsFeaturizer
  • name: CountVectorsFeaturizer analyzer: “char_wb” min_ngram: 1 max_ngram: 4
  • name: DIETClassifier epochs: 100
  • name: EntitySynonymMapper
  • name: ResponseSelector epochs: 100

Configuration for Rasa Core.


  • name: MemoizationPolicy
  • name: TEDPolicy max_history: 5 epochs: 100
  • name: MappingPolicy
1 Like