Registration Fraud Detection helps businesses address account quality issues such as fake or bulk registrations during user growth. It quickly assesses the risk level of registering users by analyzing multidimensional features, including behavior profiling, risk networks, and sample comparisons. Use the service response to manage your business logic.
Version overview
To meet the risk control needs of businesses across different industries and stages, Registration Fraud Detection offers a Basic Edition and an Enhanced Edition. The following table compares their differences:
Features | Basic Edition | Enhanced Edition |
Real-time analysis | Supported. | Supported. |
Service return value | Quantified score. | Quantified score and risk feature tags. |
Device risk monitoring | Not supported. | Supported. Includes anomaly detection capabilities for emulators, multi-instance apps, modem pools, group-controlled devices, cloud phones, HOOK, and more. |
Device fingerprint | Not supported. | Supported. |
Group analysis | Not supported. | Supported. |
Log delivery to SLS | Not supported. | Supported. You can authorize log delivery and store logs free of charge for one year. |
Service event parameters
Service event parameters are request parameters in the ServiceParameters field of common parameters, formatted as JSON. The following table lists the request parameters required for the Registration Fraud Detection service (including both Basic and Enhanced Editions).
To maximize compatibility with various input scenarios, Alibaba Cloud Fraud Detection does not validate the business format of string inputs. Validate the business format of your data yourself. For example, for the mobile field, verify that the input matches the standard mobile phone number format used in the Chinese mainland (an 11-digit numeric string starting with 1).
Field name | Supported edition | Description | Data format | Example | Required |
accountId | Basic Edition, Enhanced Edition | Account ID that uniquely identifies an account. | String | 10123**** | No (recommended) |
operateTime | Basic Edition, Enhanced Edition | Operation timestamp in seconds, in Greenwich Mean Time (GMT). Note When scanning historical data for risk, provide the actual historical operation time to avoid misidentification due to incorrect timing. | Long | For example, GMT 2018-04-01 12:00:00 corresponds to timestamp 1522555200. | No |
mobile | Basic Edition, Enhanced Edition | Include the country/region code in the format | String | 001-718123**** | No (recommended) |
ip | Basic Edition, Enhanced Edition | Public IPv4 address of the client (user side) at the time of the event. | String | 42.120.XX.XX | No |
Basic Edition, Enhanced Edition | Email address. | String | admin****@aliyun.com | Yes | |
deviceToken | Enhanced Edition | Device token obtained using the Device Risk SDK. | String | MzQvo1d7scyZ3tl_RcJZo_QOytAjy1LWRRLoRKo5oZSoo_JGj1ZoR5JGoRo5jcdn57gV5kxVRcLER5RQoZSvRZZQRcROjcMW5csZR_RGy_55RKJ_oooqZ7dSV5gRnKxOV7eWVQQjRtlRQoAjRcM0 | No (recommended for better results) |
nickName | Basic Edition, Enhanced Edition | Account nickname. | String | admin**** | No |
userAgent | Basic Edition, Enhanced Edition | User-Agent field from the HTTP request header. | String | Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.84 Safari/537.36 | No (recommended) |
refer | Basic Edition, Enhanced Edition | Referer field from the HTTP request header. | String | https://www.aliyun.com/ | No (recommended) |
mac | Basic Edition, Enhanced Edition | Device MAC address. | String | C0:77:36:2E:XX:XX | No (recommended) |
operateSource | Basic Edition, Enhanced Edition | Source of the operation. Valid values:
| String | PC | No (recommended) |
appVersion | Basic Edition, Enhanced Edition | App version number. | String | 1.0 | No (recommended) |
deviceType | Basic Edition, Enhanced Edition | Device type. Valid values:
| String | PC | No (recommended) |
Response parameters
The Registration Fraud Detection service returns a score and risk tags. Risk tags are only available in the Enhanced Edition.
Use the following table to interpret the business meaning of the score field in the Data response parameter and apply appropriate actions based on recommendations from Alibaba Cloud's risk control team.
In practice, you can choose flexibly based on your business operations.
Score range | Risk level | Action recommendation |
85–100 (inclusive) | High risk | Apply strong verification (such as SMS verification or ID Verification) or restrict access to high-risk services. |
65–85 (65 inclusive, 85 exclusive) | Medium-high risk | Apply moderate verification (such as Captcha, ID Verification, or SMS verification). |
35–65 (35 inclusive, 65 exclusive) | Medium risk | Apply simple verification (such as SMS verification or security questions). |
0–35 (0 inclusive, 35 exclusive) | Low risk | Allow the request or monitor with tagging. |
Contact Alibaba Cloud support for details about the business meaning of the tags field in the Data response parameter.
For more information about common parameters, see Common parameters.