Transactional Intelligence

Transactional Intelligence

You can enhance the Inbenta Chatbot with additional functionality to integrate with CRM, billings systems and virtually any software needed to accomplish the task of the Chatbot. This is done with a tool called Webhook.

Chatbot Highlights

  • Unique Natural Language Processing engine with over ten years of lexical development
  • Works out of the box: no additional rephrasing or learning necessary
  • Webhook-based Decision Trees for a dynamic, context-sensitive conversational flow
  • Multiple entity-based detection method that interpret visitors’ intents
  • Scalable and expandable solution that integrates with CRMs, billing systems, and more...
  • Feature-rich transactional intelligence with seamless escalation to live conversation
  • Unique avatar options for a deeply personalized customer experience

A Webhook is an HTTP callback that can be used to trigger transactions and back office calls to any external systems. It can also determine the flow of Decision Trees, assign values to Variables, and collect additional information about the user and the context.

It is also possible to implement Webhooks as JavaScript callbacks. This way, the Inbenta Chatbot can benefit from an encrypted connection from any browser with SSL, which makes sure that communications remain secure.

Occasionally, the Inbenta Chatbot might need additional information to fulfill its tasks. This information can come from different sources:

  • the user can provide it in their initial sentence,
  • the bot might ask the user for it,
  • the bot can retrieve it from back-office transactions, or
  • the information might be part of the Chatbot’s execution context.

“Entities” are the detection method that is used to capture the relevant information and store it in a variable. Inbenta supports many detection methods, ranging from simple string matches (with or without a spellcheck) to regular expressions, to complex Natural Language Patterns that allow bot masters to easily capture any relevant piece of information.

For example, it is easy to create an intent like “Purchase Flight” that uses entities: you define variables with DepartureDate, ReturnDate, OriginAirport, DestinationAirport, and link these to Inbenta’s system-defined entities to create the match.

Examples:

“I need to fly to NYC tomorrow”
“Buy a flight from Barcelona to San Francisco on 4/2”
“Buy a ticket to Berlin”

All these queries will match the appropriate intent. If entities are missing, the Inbenta Chatbot will ask for the necessary variable values interactively.

You can also use Webhooks to assign values to variables and entities. The Inbenta Chatbot is an extremely flexible tool that can adapt to any use case.