DataVisor, the award-winning AI-powered fraud management company, announced the launch of DCube. DCube builds on DataVisor’s pioneering fraud platform to deliver a single, turnkey solution that introduces unprecedented advances to the current fraud management landscape.
DCube supports the fraud management needs of large enterprises by enabling organizations to minimize risk exposure, while putting control in the hands of its users to proactively detect and stop fraud in real time.
“As the enterprise rapidly moves towards digitalization, predicting and preventing fraud is critical in order to improve customer experience and minimize risk while ensuring compliance,” said Yinglian Xie, CEO and co-founder of DataVisor.
“DCube is built to manage complex digital signals and behavior analytics at big data scale, and we put this technology directly into the hands of our users. At DataVisor, the essence of our platform is built on unsupervised machine learning, which, when coupled with traditional machine learning capabilities, can quickly adapt and detect new and unknown types of fraud attacks. DCube will not only keep our clients one step ahead of malicious actors, but also provide them with greater control.”
With DCube, large enterprises can react to sophisticated fraud quickly and with greater agility. DCube provides data scientists and fraud teams with model transparency and configurability, and a proprietary unsupervised machine learning (UML) engine that can detect unknown fraud early-on, without the need for training data or labels.
DCube users have the ability to build and tune their own models, enabling them to experiment and test the technology on various input data and use cases.
DCube features a hyper-modern architecture built to manage complex digital signals and behavior analytics using the most advanced machine learning technologies at big data scale, empowering large enterprises to identify and prevent even the most sophisticated attacks.
DCube provides advanced capabilities for:
- Data validation and integration: Data Scientists can accelerate fraud model development and increase accuracy by establishing data quality early. Once a user uploads their data and maps the fields, DCube automatically analyzes quality and flags potential issues; the system also seamlessly integrates data from multiple sources.
- Feature engineering and model development: Data Scientists benefit from an extensive library of the most sophisticated and advanced out-of-box fraud features and automatic model building capabilities, and can additionally engineer custom features for optimal model performance in a fully automated manner.
- Model validation and deployment: Fraud and data science teams can review detection results, compare models, improve performance, and deploy in production for significantly enhanced operational efficiency.
- Decision models and workflows: Fraud managers and strategists can create advanced decision models and workflows, consolidate and assemble third-party signals and detection results from machine learning models and rules engines to make customized intelligent decisions and actions.
- Case management: Fraud teams can efficiently review cases, take bulk action on groups of correlated accounts, and set auto-decisioning rules to improve review speed by orders of magnitude. They can access a full audit trail to view historical activity on the account, including who performed the actions, and for what reasons.
- Advanced analytics: Fraud and Risk teams can deconstruct the events within fraud attacks to deliver actionable insights alongside precision and recall analysis. They can investigate complicated cases with detailed detection reason codes and view correlated activities across all accounts.
In an era in which fraudsters are constantly evolving their techniques and mounting ever-more sophisticated and coordinated attacks, fraud and risk management teams face a critical challenge.
Business pressures demand they respond fast, and in real time, to mitigate online threats without compromising user experience. At the same time, they must address the practical realities of rising operational overhead and increasing regulatory scrutiny and compliance requirements.
DCube offers a comprehensive solution that addresses the needs of modern fraud teams and data scientists, by enabling them to harness the power of AI, and leverage transformational supervised and unsupervised machine learning techniques to offer seamless customer experiences.