下载PDF
Software AG > 实例探究 > Squashing Financial Fraud Faster with the Power of Predictive Analytics
Software AG Logo

Squashing Financial Fraud Faster with the Power of Predictive Analytics

技术
  • 分析与建模 - 预测分析
适用行业
  • 金融与保险
适用功能
  • 商业运营
用例
  • 欺诈识别
  • 质量预测分析
服务
  • 数据科学服务
挑战
The customer, a leading global financial services company, was facing challenges due to its diverse portfolio of services and management options. The complexity of analytic models increased as the company grew, and the internal data science and IT teams were reaching their capacity. This increased the underlying risk of the company’s business model and threatened to remove a competitive advantage. The company was using a dedicated team of data scientists creating hand-coded fraud models. However, with millions of customer accounts, a large service portfolio, new product launches, and geographically dispersed operations, manual coding became a major liability. The process of converting algorithmic fraud models to “production ready” dramatically slowed the process of integrating them into the operational business processes.
关于客户
The customer is a leading global financial services company providing data and analytics solutions to businesses, consumers, and governments. Its solutions range from all aspects of credit modeling and scoring, to fraud detection and prevention, to identity management and verification services. The company offers customers a diverse portfolio of services and management options. With millions of customer accounts, a growing and large service portfolio, new product launches, and geographically dispersed operations, the company was facing challenges in managing the complexity of its operations. The company's annual revenue exceeds $3 billion a year with an operating income of more than $2 billion.
解决方案
The company implemented Zementis Predictive Analytics, part of the Software AG Digital Business Platform. This solution combined machine learning, artificial intelligence technologies, and next-generation Internet of Things-type streaming data analytics to provide automated predictive models for better risk scoring and fraud detection. The initial implementation focused on detecting anomalies in financial transfers, with the goal of identifying money laundering. With strong results, the company adopted Zementis Predictive Analytics more broadly in its cross-channel fraud detection efforts. The solution provided its functionality as a plug-in tool for other leading analytics and data warehouse platforms, using the Predictive Model Markup Language (PMML) industry standard.
运营影响
  • The company was able to automate the process of operationalizing fraud management models without the need to manually write custom code.
  • The company's data scientists were able to focus on modeling fraud rather than fixing mistakes.
  • The company was able to use advanced predictive models to detect and analyze customer behavior, market dynamics, security risks, and other variables necessary for quick and precise fraud detection.
数量效益
  • Analytics-based decisions that previously required months of preliminary analysis now required only days, sometimes even hours, enabling near real-time decision making.

相关案例.

联系我们

欢迎与我们交流!

* Required
* Required
* Required
* Invalid email address
提交此表单,即表示您同意 IoT ONE 可以与您联系并分享洞察和营销信息。
不,谢谢,我不想收到来自 IoT ONE 的任何营销电子邮件。
提交

Thank you for your message!
We will contact you soon.