Page History

This recognizer uses a frozen Universal Sentence Encoder TensorFlow model to encode (using sentence embedding vectors) a list of Frequently Asked Questions. Then tags sentences that match a question/answer pair given a specified threshold of accuracy with the question/answer from the FAQ. Most recently the FAQ also can use Saga Python Bridge to encode the sentence with different algorithms.

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

Library: saga-fag-stage

Configuration Parameters

• Parameter
  summary Name of the resource containing the questions of the FAQ name questions required true

• Parameter
  summary Indicates the tag which is going to be use for tagging default questionTag name tagWith

• Parameter
  summary Path to the TensorFlow model in case the TensorFlow model is going to be used name modelPath required true

• Parameter
  summary Threshold indicating when a predictions is acceptable default 0.6 name threshold type double

• Parameter
  summary Indicates if the FAQ is going to use TensorFlow or Saga Bridge default true name useTensorFlow type boolean

• Parameter
  summary Indicates if the sentence must be compared against the answer as well name evalAnswers type boolean

• Parameter
  summary Model name registered in the python bridge default bert-base-uncased name modelName required true

• Parameter
  summary Model version registered in the python wrapper to query default 1 name version

• Parameter
  summary Python server communication hostname default localhost name hostname

• Parameter
  summary Python server communication port default 5000 name port type integer

"questions": "saga_provider:saga_faq",
"tagWith": "question",
"modelPath": "./tf-model",
"threshold": 0.6,
"useTensorFlow": true,
"evalAnswers": true,
"modelName": "bert-base-uncased",
"version": "1",
"hostname": "localhost",
"port": 5000,

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?]-^ 
^----------------------------[{question}]---------------------------^

Output Flags

Lex-Item Flags:

• SEMANTIC_TAG - Identifies all lexical items which are semantic tags.
• ANSWER- Identifies a lexical item holding an answer from the FAQ.

Vertex Flags:

Resource Data

Description of resource.

Resource Format

"_id" : "KGAAJGsBemSwA0nZTLXA",
"tag": "animals",
"question" : "This is a cat?",
"answer" : "That's a tiger",
"fields" : {
    "questionVector" : [ 0.058885570615530014, -0.002637901809066534, ... ],
    "answerVector" : [ 0.34535465645635530014, 0.002342342343453453, ... ],
    "url" : "https://isthisacat.com"
    . . . additional fields as needed go here . . . 
}
"confAdjust": 0.95

Fields

• Parameter
  summary question you want to compare against the sentence name question required true

• Parameter
  summary answer of the question, that can be also be compare against the sentence name answer required true

• Parameter
  summary metadata of the FAQ entry name fields type json required true

  
  

  • Parameter
    summary context vector of the question, size can change depending of the algorithm used to build it name questionVector type double array required true

  • Parameter
    summary context vector of the answer, size can change depending of the algorithm used to build it name answerVector type double array required true

  • Parameter
    summary for use in external applications name url

• Parameter
  summary Tag which will identify any match in the graph, as an interpretation name tag required true

  • These will all be added to the interpretation graph with the SEMANTIC_TAG flag.

    Tip
    Tags are hierarchical representations of the same intent. For example, {city} → {administrative-area} → {geographical-area}