NAV Navbar
shell

Getting Ready

API Server

api.allganize.ai is Allganize's NLU API end point. If you are using on-prem, please use your API endpoint.

Authentication

The NLU API uses API keys to authenticate requests. If you don't have them yet, please contact Allganize.

Named Entity Recognition API

Named Entity Recognition API extracts the important keywords from natural language texts in your domain.

Endpoint

curl https://api.allganize.ai/v1/ner \
-d text="I am traveling from Munich to Paris by Air France business class" \
--header "API-KEY: YOUR_API_KEY"

POST /v1/ner

Input Data

Parameter Description
text input texts you want to analyze. English, Japanese, Korean and Chinese are supported.

Response

{
  "result": [
    { "start": 20, "end": 26, "tag": { "name": "DEPARTURE" }, "token": "Munich" },
    { "start": 30, "end": 35, "tag": { "name": "ARRIVAL" }, "token": "Paris" },
    { "start": 39, "end": 49, "tag": { "name": "AIRLINE" }, "token": "Air France" },
    { "start": 50, "end": 64, "tag": { "name": "SEAT_CLASS" }, "token": "business class" }
  ]
}

The response has a list of inferred results for the requested text.

Text Classification API

Find out what User intent is and formulate the best response and actions. Our Text Classification automatically identifies user intents. NLU model can be further trained and optimized to your custom intents.

Endpoint

curl https://api.allganize.ai/v1/classification \
-d text="He is a male. And name is Matt." \
--header "API-KEY: YOUR_API_KEY"

POST /v1/classification

Input Data

Parameter Description
text input texts you want to analyze.

Response

{
  "result": {"intent": "PROVIDING_INFO"}
}

The response has a classified result for the requested text.

Sentiment Analysis API

Sentiment analyzer uses machine learning to reveal the structure and meaning of text. You can extract information about people, places, and events, and better understand customer sentiment.

Endpoint

curl https://api.allganize.ai/v1/sentiment \
-d text="Loved the location and the breakfast. Would definitely stay here again!" \
--header "API-KEY: YOUR_API_KEY"

POST /v1/sentiment

Input Data

Parameter Description
text input texts you want to analyze.

Response

{
  "result": {"sentiment": "like"}
}

The response has a classified result for the requested text. (like or dislike)

Question and Answering API

Question and Answering API understand requested text and automatically responds from the questions and answers you provide.

Endpoint

curl https://api.allganize.ai/v1/qa \
-d text="Can I cancel my order?" \
--header "API-KEY: YOUR_API_KEY"

POST /v1/qa

Input Data

Parameter Description
text input texts you want to analyze.

Response

{
  "result": [
    {
      "question": "How can I cancel my order?",
      "answer": "You can cancel an online order within 30 minutes of placing the order. Sign into oyour account and looking up the order number and click cancel on the order details page."
    }
  ]
}

The response has a list of the pairs of question and answer.

Machine Reading Comprehension API

MRC API finds a single answer directly within the text documents you uploaded.

Endpoint

curl https://api.allganize.ai/v1/mrc \
-d text="Who published the article titled Computing Machinery and Intelligence?" \
--header "API-KEY: YOUR_API_KEY"

POST /v1/mrc

Input Data

Parameter Description
text input texts you want to analyze.

Response

{
  "result":"In 1950, Alan Turing published an article titled \"Computing Machinery and Intelligence\" which proposed what is now called the Turing test as a criterion of intelligence." 
}

The response is the paragraph containing the answer.

Review Conversion API

Review Analyzer identifies recurring themes and topics in comments and reivews. Allows retailers to understand how their customers feel without having to read every reviews individually.

Endpoint

curl https://api.allganize.ai/v1/review \
-d text="Cannot recommend highly enough! Absolutely fantastic apartment in a fantastic area! Great thoughtful touches and close amenities! Good vibes all round!" \
--header "API-KEY: YOUR_API_KEY"

POST /v1/review

Input Data

Parameter Description
text input texts you want to analyze.

Response

{
  "result": [
    [
      { "category": "location", "sentiment": "2" },
      { "category": "cleanliness", "sentiment": "4" },
      { "category": "rooms", "sentiment": "3" },
      { "category": "sleep quality", "sentiment": "4" },
      { "category": "service", "sentiment": "1" }
    ]
  ]
}

The response contains a list of sentiment of each categories.

Summarization API

TBD

References

Tools & Library

GraphQL is implemented over standard HTTP and websocket. So you can choose any http, websocket library like python requests.