Offline Text Result Acquisition API
API Address
https://as.dun.163yun.com/v3/text/callback/results
API Description
Online detection of suspected data in the API results will be performed offline and manually. The offline results of these data require the client to obtain periodically through the API. After getting the result of data processed, the next call will not return the offline data obtained before. The API limits the frequency of requests. If the request frequency is too fast, the server will refuse to process. It is recommended to obtain it once every 30 seconds.
Request parameter
The API parameters are consistent with the public request parameters. For details, see Public Request Parameters
Response Results
The response field is as follows, the common response field has been omitted, see Response General Fields:
Result array data structure
| Parameter Name | Type | Description |
|---|---|---|
| action | Number | Test results, 0: Pass, 2: Not pass |
| censorSource | Number | Source of audit, 0: Yi Dun human audit, 1: customer audit, 2: Yi Dun machine audit, version support V3.1 or above |
| censorRound | Number | Number of audit rounds: 1:1st round, 2:2nd round, 3:3rd round, 4:4th round, 5:5th round. No audit mode is set, and the default return is 0 |
| taskId | String | The data request identifier returned by the online API detection, which can query the latest result of the data according to the identifier |
| callback | String | The callback field data passed by the product call online text detection. |
| labels | jjson array | classification information |
Labels data structure
| Parameter Name | Type | Description |
|---|---|---|
| label | Number | Classified information, 100: porn, 200: advertising, 300:terrorism，400: contraband, 500: political, 600: abuses, 700: meaningless |
| level | Number | Classification level, 1: Uncertain, 2: OK |
| details | json object | Other information |
Details data structure
| Parameter Name | Type | Description |
|---|---|---|
| hint | jjson array | clue information for locating the problematic part of the text, assisting with manual review |
censorLabels data structure
| Parameter Name | Type | Description |
|---|---|---|
| code | Number | A unique tag ID is generated after the smart system has configured the custom tag |
Request example
Response example
When the offline result is obtained, the output example is as follows：
{
"code": 200,
"msg": "ok",
"result": [
{
"action": 2,
"censorSource": 0,
"censorRound":0,
"taskId": "a9840b389d2e483d9349cd62e9cda250",
"callback": "ebfcad1c-dba1-490c-b4de-e784c2691768",
"labels": [
{
"label": 600,
"level": 2,
"details": {
"hint": [
"xxx"
]
}
}
]
},
{
"action": 1,
"taskId": "11140b389d2e483d9349cd62e9cdas11",
"callback": "asdffcad1c-dba1-490c-b4de-e784c269112",
"censorLabels": [
{
"code":Customize the tag mapping ID
}
]
}
]
}
When there is no latest offline result to be obtained, the output example is as follows：
{
"code": 200,
"msg": "ok",
"result":[]
}
