Conversation Analysis
This diagram element is used to invoke a large language model workflow to obtain the semantic recognition, sentiment analysis, and Q&A search results in dialog content.
Diagram Element

Parameter Description
- Node Name: name displayed on the diagram element, which can be customized.
- Description: details of a node.
- Function: check box. The semantic recognition, sentiment analysis, and Q&A search functions are provided.
- Conversation Content: request parameter for invoking the selected function. If this parameter is left empty, the IVRREQUEST.input parameter is used by default.
- Timeout Interval: timeout interval for invoking an API function, in seconds. The value ranges from 1 to 60. The value of this parameter is the smallest one among the value of the current parameter, the API invoking timeout interval, and session timeout interval configured in the system. Default value: 5
- Extended Data: request parameter for invoking the selection function. This parameter is a reserved extended field like call-associated data.
- Result Processing - Variable assignment: The content returned by the diagram element is stored in the intent variable TOC.DialogAnalysis. Variable assignment is supported. The value entered on the right is assigned to the variable entered on the left.
- Result Processing - Condition branch: defines the branches of the current diagram element exit. The options are Simple Expression and Complex Expression. Simple expressions use the following relational operators to return a value of the Boolean type (true or false): >, >=, <, <=, ==, and !=.
Figure 1 Simple Expression and Complex Expression
Using the Diagram Element
This diagram element is used together with agent assistant services.
Typical Application Scenario
For details about how to use the conversation analysis diagram element, see the archiving flow in Intelligent Agent Assistant Flow Description.
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.
For any further questions, feel free to contact us through the chatbot.
Chatbot