dialoguekit.nlg.template_from_training_data

Extract user response utterance templates from annotated training data.

Functions

build_template_from_instances(...)

Builds the NLG template.

extract_utterance_template(...)

Extracts utterance templates for each intent.

Module Contents

dialoguekit.nlg.template_from_training_data.build_template_from_instances(utterances: List[dialoguekit.core.annotated_utterance.AnnotatedUtterance]) → Dict[dialoguekit.core.intent.Intent, List[dialoguekit.core.annotated_utterance.AnnotatedUtterance]]

Builds the NLG template.

The Intent the Utterance comes with will be used. If no intent is present for an utterance it will be skipped.

Parameters:

utterances – List of AnnotatedUtterance-s.

Returns:

Dict with Intents and lists with corresponding AnnotatedUtterances.

dialoguekit.nlg.template_from_training_data.extract_utterance_template(annotated_dialogue_file: str, participant_to_learn: str = 'USER', satisfaction_classifier: None | dialoguekit.nlu.models.satisfaction_classifier.SatisfactionClassifier | None = None) → Dict[dialoguekit.core.intent.Intent, List[dialoguekit.core.annotated_utterance.AnnotatedUtterance]]

Extracts utterance templates for each intent.

If a Satisfaction Classifier is provided it will be used to classify the utterances. The classification logic is as follows:

  • Hold participant utterance.

  • Hold counter-participant utterance.

  • Concatenate participant and counter-participant utterance and classify satisfaction.

  • The next utterance from participant will be given the satisfaction from the concatenated utterance from the previous utterances. reflecting the satisfaction at that given point in time.

Parameters:

  • Annotated_dialog_file – annotated dialogue json file.

  • participant_to_learn – Which participant we want to create a template on.

  • satisfaction_classifier – SatisfactionClassifier

Returns:

Dict with Intents and lists with corresponding AnnotatedUtterances.