IBM Watson Inbound Message Sentiment

Every inbound message sent to is analyzed for the subscriber's sentiment. It is included in the service.

Sentiment is the attitude, opinion or feeling toward something, such as a person, organization, product or location. IBM Watson Message Sentiment provides easy-to-use mechanisms to identify the positive or negative sentiment within a text message. The algorithm looks for words that carry a positive or negative connotation. It understands negations (i.e. "this car is good" vs. "this car is not good") and modifiers (i.e. "this car is good" vs. "this car is really good"). IBM Watson Message Sentiment can be used to track the voice of the customer for an organization or to support user feedback and surveys. Developers can now gain deep customer insights by making use of the vast amount of unstructured data accumulated through messages.

Read more about it from IBM!

In most cases, the sentiment score will be neutral as Watson does not have enough data to decipher individual keywords or replies from an opt-in confirmation.

However, for developers building SMS inside of:

  • CRMs
  • Customer Service Apps
  • Survey Apps
  • Political Constituency Apps

Watson can be of service to help flag both negative and positive consumer experiences for escalation or follow-up.

In the inbound message posted to your callback, there are two fields related to Watson. They are the type and the score. The type can either be, Mixed, Postive, Negative or Neutral. The score will be between (-1)-1, with the closer to 1 being positive.

Sample Inbound Message Payload

    "MessageID": "999323239129",
    "MobileID":  "+19993338888",
    "AccountID": "ca3121",
    "StoreName": "Jim's Pizza",
    "StoreID": "c23221",
    "MessageDirection": "inbound",
    "FromCountry": "US",
    "FromState": "CA",
    "FromCity": "Brisbane"
    "FromZip": "94321"
    "Keyword": "PIZZA",
    "MessageBody": "PIZZA 123",
    "SubscriberStatus": "opted-in",
    "TimeStamp": "03-22-2016 12:31:43",
    "ShongCode": "444999",
    "MessageCharacterCount": "21",
    "NumSegments": "1",
    "NumMedia": "1",
    "MediaContentType": "image/jpeg',
    "MediaUrl": "https:\/\/\/2010-04-01\/Accounts\/ACff9c",
    "Type": "Positive",
    "Score": "0.78249",