Page History

This recognizer uses a frozen Universal Sentence Encoder TensorFlow model to encode (using sentence embedding vectors) a list of intents. Then tags sentences that match an intent given a specified threshold of accuracy with the intent from the list of possible intents.

Operates On:  Lexical Items with TOKEN and possibly other flags as specified below.

Library: saga-intent-stage

Configuration Parameters

• Parameter
  summary Switch between Neural Network using Python Bridge or Universal Sentence Encoder using Tensorflow. default Neural Network (false) name Neural Network - Universal Sentence Encoder type boolean required true

  
  
  • Both options have their own set of configurations and change on the UI if one or the other is checked.

Configuration Parameters (Neural Network)

• Parameter
  summary Host where the Python-Bridge is running. default localhost name localhost

• Parameter
  summary Port where the Python Bridge is running. default 5000 name Port type integer

• Parameter
  summary Model to be used for the intent recognizer default None name Select Model required true

• Parameter
  summary Version to be used for the recognizer. default None name Select Version required true

• Parameter
  summary Percentage of similarity to be accepted by the recognizer. A higher threshold would mean closer results but less similarities found, a lower threshold would find more similarities but less closer results. default 0.6 name Match Threshold type double required true

• Parameter
  summary List of tags to be normalized by the recognizer. default None name Normalize Tags type string array

• Parameter
  summary List of tags to be removed by the recognizer. default None name Remove Tags type string array

Configuration Parameters (Universal Sentence Encoder)

• Parameter
  summary Percentage of similarity to be accepted by the recognizer. A higher threshold would mean closer results but less similarities found, a lower threshold would find more similarities but less closer results. default 0.6 name Match Threshold type double required true

• Parameter
  summary List of tags to be normalized by the recognizer. default None name Normalize Tags type string array

• Parameter
  summary List of tags to be removed by the recognizer. default None name Remove Tags type string array

"atLeastOneFlag": []
"boundaryFlags": ["SENTENCE_SPLIT", "TEXT_BLOCK_SPLIT"]
"confidenceAdjustment": 1
"debug": false
"dontProcessFlags": []
"hostname": "localhost"
"intents": "saga-provider:saga_intent"
"modelName": "bert-base-uncased"
"modelPath": "tf-models/use/"
"normalizeTags": []
"port": 5000
"removeTags": []
"requiredFlags": []
"skipFlags": []
"threshold": 0.6
"useTensorFlow": false
"version": "1"
"_encrypt": []

Example Output

V---------------------[Any analytics available for Internal Sites? Why are dogs always hungry?]---------------------V 
^----------[Any analytics available for Internal Sites? ]-----------V---------[Why are dogs always hungry?]---------^ 
^-[Any]-V-[analytics]-V-[available]-V-[for]-V-[Internal]-V-[Sites?]-^-[Why]-V-[are]-V-[dogs]-V-[always]-V-[hungry?]-^ 
^----------------------[{internalSiteIntent}]-----------------------^ 

Output Flags

Lex-Item Flags:

• SEMANTIC_TAG - Identifies all lexical items which are semantic tags.
• INTENT - Placed on all the tokens which passed the minimal threshold.

Vertex Flags: