The share of digital payments is rapidly increasing, from 8% of the overall global retail payments market in 2015 to an estimated 18–24% by 2020. So is payment fraud, which has almost tripled (with regard to the cost of fraud as a percentage of merchant revenues) between 2013 and 2016. Payment fraud comes in many forms, most notably identity and account theft, friendly fraud ,and merchant fraud.
Conventional approaches for fraud scoring, deceptive merchant detection, and merchant compromise detection often include a mix of static rules engines, state machines, big data algorithms, graph algorithms, and classical machine learning but are expensive and too slow to adapt to new fraud patterns in timely manner. Advanced machine learning techniques, including deep learning and self-learning algorithms, offer the ability to not only anticipate anomalous behaviors at the merchant and consumer levels but also deploy corresponding countermeasures dynamically and instantaneously.
Markus Kirchberg explains how recent advances in AI and machine learning, decision sciences, and network sciences are driving the development of next-generation payment fraud capabilities for fraud scoring, deceptive merchant detection, and merchant compromise detection.
Markus Kirchberg is CEO of Wismut Labs Pte. Ltd., where he leads a team of diverse technical experts that help modernize and transform clients’ products, services, and operations. Markus has over 20 years experience in research and technology-driven innovation, and his career spans academia, dedicated research centers, and industrial research and incubation labs. Previously, Markus was the head of technology innovation at Deep Labs, where he was responsible for driving and delivering technology innovation across the Asia Pacific region; headed Visa Labs, Asia Pacific; served as an expert at HP Labs Singapore, where he led various innovation initiatives on next generation, cross-domain data analytics platforms; worked as a research fellow and principal investigator at the Institute for Infocomm Research at A*STAR; and was a lecturer at Massey University, New Zealand. Markus’s skill set includes full innovation lifecycle management, automating infrastructure, cloud computing, data management at multipetabyte scale, data privacy, deep learning, emerging technologies, the internet of things, large-scale data analytics, and extreme transaction processing. He has extensive experience in healthcare, logistics, payments analytics and processing, risk management, and transportation.
©2017, O'Reilly Media, Inc. • (800) 889-8969 or (707) 827-7019 • Monday-Friday 7:30am-5pm PT • All trademarks and registered trademarks appearing on oreilly.com are the property of their respective owners. • firstname.lastname@example.org