Lake AI Service
Lake AI Service
- Documentation
Lake AI ServiceTextText security detectionText content risk identification
Copy page
Download PDFCopy pageDownload PDFPrerequisites
Operator usage prerequisites: Activate the Business Risk Identification product - Text Risk Identification.
Operator parameters
Input
Input column name | Description |
|---|---|
text_col | Input text column |
account_id_col | (Optional) Column for the sender's ID |
operate_time_col | (Optional) Column for the second-level timestamp when the user sends the text |
nick_name_col | (Optional) Column for the sender's nickname |
signature_col | (Optional) Column for the user's personal signature |
text_type_col | (Optional) Column for the type of text content (for example, "prompt", "response") |
session_id_col | (Optional) Column for the conversation round ID in AIGC dialogue scenarios |
Output
An array of structs containing risk identification results
- FinalLabel: Final risk label
- Decision: Decision result (PASS, BLOCK, REVIEW)
- Message: Information returned by the server
- risk_result: Complete detection result in JSON format
Parameters
If a parameter does not have a default value, it is required
Parameter name | Type | Default value | Description |
|---|---|---|---|
app_id | int | Application ID activated in the Business Risk Identification product | |
biztype | str | Scenario configured in the Business Risk Identification product | |
timeout | int | 120 | Timeout setting, unit (s). Sets the timeout from sending the request to receiving the result. For offline tasks, to prevent data processing from being blocked, this value can be appropriately increased. |
Examples
The following code demonstrates how to use Daft (for distributed scenarios) to run the operator for text content risk identification.
