Building Agents
Extracting Data
Extract structured data from conversations
Structured data can be extracted from the agent to be used to filter and analyze calls or to enhance post-call automatons. A common example is to classify a call into various predefined labels.
The description of the property is important as the AI will use this description to extract the data. The description can be treated as a small prompt instructing the AI on how and what to extract.
The type will determine the format of the extracted data.
Define data to extract
Navigate to theAnalysis tab in your agent configuration to define what structured data to extract. From there, you can define properties to be extracted.
The description of the property is important as the AI will use this description to extract the data. The description can be treated as a small prompt instructing the AI on how and what to extract.
The type will determine the format of the extracted data.