Have you ever wondered why your website is bearing the loss for data breaches happening on other platforms? In the face of credential stuffing attacks, you need a powerful device identification and real-time risk detection tool to distinguish between good customers and risky ones without actually seeing the users.

Veri-id is an online real-time risk detection service, which can infer and identify visiting devices and further apply to fraud detection. Veri-id helps your website identify suspicious devices and abnormal behaviors by collecting and analyzing different characteristics of devices, such as hardware and software, operating system, IP address, user operation mode, etc., to effectively prevent fraud and cyber-attacks.

Veri-id recognizes users through their devices, and can be applied to a variety of online scenarios, including account login, account personal information changes, online transfers, online credit card transactions, etc.

Product features

• Device Fingerprint Identification

Veri-id device fingerprints are inferred through HiTRUST's patented algorithm from nearly a hundred devices information and operating characteristics. The collected device information includes the device model, operating system version, system configuration, mouse movement track and keyboard typing rhythm, etc. The combination of these attributes can form a device's fingerprint, just like a human fingerprint. It is a way to identify a device without interfering the user experience. A correct recognition of each device is crucial to past records accumulation and abnormal access detection, such as whether the same device sends a large number of requests in a short period of time, whether the account is used on an unfamiliar device, whether the device configuration and historical habits used by the account are the same, etc.

• Explainable AI Models

Through machine learning technology, Veri-id gains insight into customer behavior patterns in the ever-growing data and predicts the probability of fraud in advance. Different from other AI products on the market, we model different fraud attack methods (such as fraudulent devices, Account Takeover, Card Testing/Account Testing, etc.), and provide reasons to explain for the risk of model derivation. We make the black-box AI model transparent, allowing users to apply it more effectively and trust the model prediction results.

• Rule Engine

The flexible architecture and friendly user interface make it easy for non-IT personnel to operate. In addition to building rules to protect against known methods of fraud attacks, risk scores from AI models can be used as input to the rule engine during rule creation to optimize the strengths of different tools.

• Customized Decision-making Path

Define follow-up actions for each risk level based on internal business strategies and decide when user interaction should be involved. The decision-making path can be adjusted at any time, taking effect immediately.

• Data Visualization

Store and perform statistical calculations through ElasticSearch, analyzing transactions and device information in real time. Use Kibana to present massive data in the most appropriate charts to make data exploration more efficient.