Fraud crosses many use cases and affects multiple industries. These use cases are interlinked and sometimes cross each other, including: AML, transactions & credit card, cyber & malware, financial crimes, insider threat, and compliance & audit. It is well established that graph algorithms and Machine Learning (ML) are useful for calculating similarity and identifying communities for the most common fraud use cases, but there are new streaming techniques that combine time-series streaming for better results.
The focus of this webinar is to identify how the Imply Technology in conjunction with Machine Learning, Visualization and new technology can directly increase the accuracy and output of systems and how including the ‘Human in the Loop’ can get you ahead of fraud. This event is designed as a 'Speed Dating' format, focusing on key topics for under 15 minutes in order to maximize the insights. During this online meetup, you'll learn from our experts how Expero and Imply can unlock the potential in your organization. We will feature unique Expero lightning talks on ML & Business Visualization technology, followed by a short Q&A session.
Key Learning Topics:
- What is the status quo in current fraud detection approaches: See how a combination of graph and machine learning techniques can identify and segment cases worthy of further scrutiny
- How streaming techniques can improve results: Illustrate how Imply’s technology in conjunction with existing ML & Graph can further improve anomaly detection
- Methods to reduce false positives: Review how Imply, ML & Graph combination techniques reduce false positive signals
- How Imply’s technology is uniquely positioned for financial services anomaly detection: See how massively scalable, real time streaming analytics enables quick training of ML models such that fraud detection schemes may be deployed more rapidly and real time dashboards from Imply can be used for monitoring
- Use of Expero’s Visualization for Suspected Fraud Case Inspection: See how a fraud workbench visualization can benefit from mixed graph and stream processing
Session 1 (10 min): Why fraud use cases require real time, ML, and graph technology. Why Imply and why now?
Session 2 (20 min): Identify visualization & Human interaction with fraud. Discuss new approaches and 'The Art of the Possible' Fraud Workflow Demonstrations - Management Investigations and ML & Data practitioner views
Session 3 (20 min): A review of how Imply’s platform can alter the current approach and how it can be applied to fraud detection, & decrease false positives in a real time environment
Session 4 (5 min): Tie together the different elements of ML, Visualization & Technology with the Imply Technology for a cohesive approach to Fraud identification, analysis and prevention
Session 5 (5 min): Open Q&A