Natural language processing text analyzers
When a Pega Customer Service application receives incoming messages from the Facebook Messenger channel, these messages are processed through a series of text analyzers to determine what a customer wants to accomplish. These text analyzers are processed in a sequential order.
Pega Customer Service provides the following text analyzers for Pega Intelligent virtual assistant for Facebook Messenger.
||This analyzer routes incoming messages to an escalated chat session.
||This case-insensitive text analyzer interprets the exact text that is entered in the message and compares it to the literal names of commands.
||This analyzer supports dynamic internal runtime behavior, such as chat escalation and messages that originate from a customer service representative.
|Pega NLP Generated
||This analyzer is used to add keywords, such as should, must, or not, for each service case and response that is configured for the channel.
||This analyzer enables you to configure specific keyword patterns, such as cancel and abort, that might be interpreted as commands during case processing.
||This analyzer is used in circumstances when the customer is not in the middle of a service case.
||This analyzer is used to queue a response from a customer when the analysis of other analyzers results in either a case or a response that needs customer authentication. If the user has not provided credentials within the specified time, then the response is queued until the authentication is completed. In addition, if the sentiment calculated is low, then the chat with agent case is started to escalate the chat to an available customer service representative.
Pre-natural language processing and post-natural language processing enable the system to handle different scenarios, such as deciding whether the customer needs to escalate the chat to an agent or determining if the customer is currently in the process of running a case flow. Natural language processing analyzers are linked to the Pega natural language processing engine.
The exact text analyzer compares the text that is entered against the list of available configured responses for an exact match. This comparison is useful for commands that you want to run, such as those that are preceded with the word _cmd. You do not need to modify any of the text analyzers except for updating the taxonomies that are related to the natural language processing analyzers.
For natural language processing configuration, you can configure either a rule-based model or free-text model. For more information about deciding which model to use, see Analyzing text samples in the NLP Sample application on the PDN.
You can use the sample rule-based model as a starting point and modify it according to your business needs. For more information about configuring a natural language processing taxonomy, see Requirements and best practices for creating a taxonomy for rule-based classification analysis on the PDN.
Published November 27, 2017 — Updated February 6, 2018