Dear friends and partners!
We are pleased to announce an important event of this year—the launch of the new, 16th generation JuicyScore API and JuicyID. In this version, we have added a range of new markers, strengthened parameters for detecting remote access and social engineering, and significantly improved the technology stack for detecting signs of randomization and virtualization on devices. We have also greatly enhanced the IDX indices to improve accuracy and security.
What’s New in API 16?
The JuicyScore 16th generation API offers updated tools to enhance the informativeness and protection of your business, as well as to improve the customer experience.
- 17 new variables have been added.
- The usefulness of scoring and the Gini coefficient has been increased by 25%+, and IV by 50%+.
- Type I and Type II errors for device authentication have been reduced—new cross-browser and cross-domain matching technologies have been introduced, and existing device matching technologies have been updated.
- Two new, significant groups of malicious randomizers and countermeasures have been identified.
- Service performance and speed have been improved by 25%+.
API 16: New Variables
The new variables expand the capabilities and improve the accuracy of fraud detection decisions. Additionally, several new variables provide enhanced capabilities for analyzing high-risk software and configurations on the applicant's device. Here are some of them:
NetFilter vector length: The number of traffic filtering/correction apps installed on the device. In the new version, we have increased the number of high-risk app groups considered on the device.
Is WebRTC data matching: A parameter that shows the correctness of the collection and retrieval of WebRTC-related data.
Is TLS randomizer: A parameter indicating the presence of TLS header randomization.
Is Limited Noise Randomizer: A parameter indicating the presence of minor artificial noise in browser API integrity checks.
Number of days from previous session: A parameter that shows the number of days since the last request from the device within accounts of the same holding/group. Now we can not only view the number of requests of various kinds from the device but also the number of days since the last request. This enhances the accuracy and informativeness of our vector and allows for more detailed flow analysis.
Number of bot beh anomalies: A parameter that indicates the number of behavioral tests triggered to detect bots. This parameter helps improve the accuracy of analysis when signs of bot technology usage are detected on the device.
Generic Score
The Generic Score is an ensemble of models built on the aggregates IDX1-IDX10 and several auxiliary variables. Below is the growth dynamics of the informativeness of the indexes in the latest versions of our API.
IDX1: Stop Markers
IDX1 consists of more than 40 rare events that have a high probability of fraud through technical manipulation of devices or the creation of significant anomalies in the internet connection. This variable covers all methods of device randomization detection, as well as techniques for interfering with their "digital fingerprint."
IDX2: User Behavior Markers
This aggregated variable combines various user behavior markers. The JuicyScore vector contains dozens of markers related to user behavior, and the main task in designing this variable is to identify stable markers. The combination of these markers into a single aggregate helps detect high-risk segments associated with high-risk device behavior, regardless of the online company’s geographic presence.
IDX3: Device Markers
IDX3 is a combination of secondary risk markers and anomalies related to the device. Each anomaly individually may indicate a potential threat, but their simultaneous manifestation highlights a high-risk area. High values of this index can serve as filters for automatic denial.
IDX4: Internet Connection Markers
IDX4 is a combination of network parameters and anomalies. High values of this variable can be used to detect and prevent fraud.
IDX10: Disposable Income Index
This index reflects the risk segment based on disposable income. The higher the value of this variable, the lower the risk. We recommend using this parameter for assessing both operational and credit risk.
The indices and variables listed above are just a small part of the improvements in API 16. We understand the importance of continuously expanding the set of markers and stop-factors that allow us to quickly reduce costs and minimize losses from fraud. The JuicyScore 16th generation API is an important step toward enhancing the security and efficiency of your business. The new features and improved parameters will provide more precise fraud protection, improve analysis quality, and enhance the customer experience.
We are confident that these updates will help you respond more quickly to threats and make better-informed decisions. Our team continues to work on developing technologies so that you always have access to the best tools to protect and grow your business. Thank you for your trust and partnership!