Credit Card Fraud Solution(1)Sift Science

最新推荐文章于 2025-08-12 14:37:12 发布

原创最新推荐文章于 2025-08-12 14:37:12 发布 · 149 阅读

0 ·

CC 4.0 BY-SA版权

文章标签：

#javascript #ruby #scala #ViewUI

Summary 专栏收录该内容

381 篇文章

订阅专栏

本文介绍了一种使用Sift Science解决方案进行信用卡欺诈检测的方法。通过集成Sift Science的JavaScript片段收集用户数据、IP地址及设备指纹信息，并通过HTTP API发送信用卡及订单详情。根据评分API返回的分数采取不同措施：高于95的视为欺诈并拒绝；低于60的接受订单；介于60到95之间的待审核。审核后将结果通过Label API反馈。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

Credit Card Fraud Solution(1)Sift Science

Commercial Solution
Sift Science
https://siftscience.com/
https://siftscience.com/image/integration/implementation-flowchart-2x.png flow

I read Sift Science documents. Here is how we do the integration. (Flow Diagram)
Step 1:
They require us to include their JavaScript Snippet to gather the user data, IP address and Device Fingerprint(Browser info and etc).

Step 2:
Then we will send the credit card information and order information, for example 6 first credit number, last 4 credit number, order items and etc thought HTTP API(they provide PHP, PYTHON, RUBY, since it is based on HTTP, we can write our own for Scala).

Step 3:
We will call their Score API to get a score: >95, it is likely a fraud. We will reject that order; <60, we will accept that order; 60<score<95, we will category the order into to-be-review and review them manually.

Step 4:
After we review the orders, we mark the order ‘fraud’ and ’not fraud’ and send this info to their Label API.

References:
https://siftscience.com/
Credit API Solution
http://www.authorize.net/