AI for Fraud Detection Virtual Summit schedule

How to Succeed with Machine Learning Against Fraud

Applying data science to predict fraud becomes an industry standard. But machine learning in fraud has its own specifics compared to other domains: highly imbalanced data, different cost of false positives and false negatives, interaction of ML models with the whole fraud system, etc. In my talk, I will share what to do with such caveats: how to choose the correct metric for model training based on business problems, how to improve models with undersampling methods, what to do in case of real-time fraud detection systems. Also, I will talk about an important business part: how to measure impact of ML models and why sometimes you need to let fraudsters in.

Elizaveta is a Senior Data Scientist at Bolt, leading European mobility platform. Her current focus is on commerce topics where she applies models and algorithms to reduce fraud and optimise payments; she is involved in the whole data science pipeline starting from problem definition and model prototyping to live ab-tests and model operations. Being passionate about math and having degrees in finance and quantitative economics, Elizaveta transitioned to Data Science from Business Analytics, marking her path with numerous math competitions and hackathons. Elizaveta is active in Data Science community in Estonia (organising and speaking at conferences and meetups). She is passionate not only about Data Science, but also about engaging more people into the tech industry (organizing Tallinn Django Girls workshop, speaking at events and camps for school kids about career in IT).

The AI Antifraud Journey in Insurance

AI can help and impact on fraud detection in many different ways.

Different approaches are determined by the different points of action inside the policy and claim process; the same can be said about different type of businesses (car, life, etc.). But, are these approaches really completely different or are there any common point?

During this brief speech we'll travel through the use of AI in the discovery process of frauders. Some of our stops will be classic and advanced applications, structured and unstructured data and the main challenges to tackle.

Valeria Verzi is currently Senior Data Scientist at Generali Italia. Valeria’s role consists of exploiting the deep technical knowledge and the business understanding to develop cutting edge artificial intelligence models to drive internal digital transformation.

She got a bachelor in Mathematical Engineering, for which she developed a thesis on multi-objective optimisation at ABB. Valeria then took the master’s degree with honors in sound and music Engineering (Computer Engineering) at Politecnico of Milan. She wrote her master thesis at Queen Mary University of London on piano inharmonicity mathematical and computer modelling. Valeria continued her career as data scientist moving to industry in 2014.

Financial Crime & Fraud Detection Using AI: Focus in Graphs: Key applications, Lessons Learned & Industry Trends

In recent years and specially during the COVID 19 pandemic, there has been an unprecedented growth in digital ecommerce that has significantly increased payments. This in turn has increased considerably fraud and financial crimes. In this recent landscape, traditional fraud detection approaches mainly based in rule-based engines presented a number of growing challenges, which both an ever growing number of cases but more importantly the number of false positives being significantly increased which has put a strain in the Fraud and Financial crime investigation teams. In this context, AI and in particular ML approaches using graphs has gained a lot of traction in the marketplace. Whilst these solutions have indeed significant potential to address the challenges of the legacy rule-based system, there are also a new type of considerations that must be observed. We will explore during the session some use cases and how to successfully addressed the main challenges implementing these graph systems. We will also explore future trends

Javier Campos is the Head of the Experian DataLabs for UK&I and EMEA. He oversees innovation with emphasis on the development of new Data-driven products and services. Pioneering application of Artificial Intelligence in mobile, voice, fraud, credit, marketing, social media, digital advertising and healthcare. Javier is also a permanent member of the Bank of England & FCA Artificial Intelligence Public-Private Forum, an industry group created to accelerate AI adoption within the Finance industry. He is a frequent speaker at industry conferences. Javier has 28+ years’ experience globally within the Finance, AI, Market Research, Media and Technology fields including serving at Global Chief Technology Officer at Kantar-WPP, Havas Media and CIO at GroupM-WPP.

Graph Analytics Applications in Agent Misconduct

Agents and brokers accounts for significant market share for insurance distribution channel and remain the key players in advising and managing insurance products. Fighting agent misconduct is therefore critical due to its scale. Nowadays, we tend to see more complicated schemes in misconduct activities. As we accumulate vast amounts of data, we observe many duplications and noise in data. Records that represent the same individuals or companies. Manually reviewing all data is costly, time-consuming, and practically impossible for large global organization. To uncover fraud patterns, it is essential to look beyond individual data points and focus more on the connections that link them With the help of advanced analytics, we can build a more effective defense solution. In this talk, we will focus on two graph analytics techniques: 1) community detection, 2) graph anomaly detection and their role in combatting agent misconduct.

Key Takeaways:

*Understand the importance of agent misconduct

*Become familiar with community detection techniques

*Become familiar with graph anomaly detection techniques

Mina is a Senior Data Scientist in Group Advanced Analytics at Manulife. Her work is focused on fraud detection using machine learning techniques including graph analytics and deep learning. In her role, the main use cases of fraud detection are customer account takeovers, new account frauds, payment and transaction frauds, and agent misconduct. Mina deals with global projects and is actively involved in designing, developing and operationalizing fraud detection applications. She previously worked with multiple business lines in banking industry and contributed to projects such as efficient pair trading, house prices index, and forecasting models for income and deposits. Mina holds a Ph.D. in Statistics and has more than 4 years of experience in data science.

Low Latency Fraud Prevention with Redis Labs

Time is money, and the difference between milliseconds and microseconds can mean the difference between big profits or losses. In modern business low latency is important, but it’s critical in banking and financial services, especially since fraud has grown to epidemic proportions. Fraudulent charges are costing companies billions of dollars, and that’s why it’s important for IT decision makers in financial services to measure latency in microseconds. A low-latency, scalable, in-memory data store like Redis can enable organizations to architect to meet this critical business need. Drawing from first principles and best practices, learn how to architect to prevent fraud in milliseconds.

Key Takeaways:

*Understand the importance of fraud prevention, especially given the rise of digitalization

*Become familiar with first principles of real-time data fraud prevention, and its associated platform components

*Appreciate the technical merits of in-memory databases such as Redis in facilitating this architecture

*Learn best practices of implementing low latency fraud prevention based on examples from the industry related to 1) e-mail fraud and 2) payment fraud

As Chief Business Development Officer, Taimur is responsible for developing emerging businesses at Redis Labs, and strategic business & corporate development. He is currently leading initiatives related to AI/ML. Prior to Redis Labs, Taimur led Worldwide Customer Success for Microsoft’s Azure Data Platform, Analytics & AI business. And before that, he was Managing Director for Amazon Web Services (AWS) Platform Technology and Applications where he led business development from 2008 (near its inception) to 2018 when the business reached $25B in ARR. Taimur helped forge some key partnerships and customers including Airbnb, CapitalOne, Dropbox, Liberty Mutual, NASA JPL, Nasdaq, Netflix, Nintendo, Intuit, SAP, Samsung, and Societe Generale.

Low Latency Fraud Prevention with Redis Labs

Time is money, and the difference between milliseconds and microseconds can mean the difference between big profits or losses. In modern business low latency is important, but it’s critical in banking and financial services, especially since fraud has grown to epidemic proportions. Fraudulent charges are costing companies billions of dollars, and that’s why it’s important for IT decision makers in financial services to measure latency in microseconds. A low-latency, scalable, in-memory data store like Redis can enable organizations to architect to meet this critical business need. Drawing from first principles and best practices, learn how to architect to prevent fraud in milliseconds.

Key Takeaways:

*Understand the importance of fraud prevention, especially given the rise of digitalization

*Become familiar with first principles of real-time data fraud prevention, and its associated platform components

*Appreciate the technical merits of in-memory databases such as Redis in facilitating this architecture

*Learn best practices of implementing low latency fraud prevention based on examples from the industry related to 1) e-mail fraud and 2) payment fraud

Sid Anand currently serves as the Chief Architect for Datazoom, where he builds autonomous streaming data systems for Datazoom's high-fidelity, low latency streaming analytics needs. Prior to joining Datazoom, Sid served as PayPal's Chief Data Engineer, focusing on ways to realize the value of PayPal's hundreds of petabytes of data. Prior to joining PayPal, he held several positions including Agari's Data Architect, a Technical Lead in Search @ LinkedIn, Netflix’s Cloud Data Architect, Etsy’s VP of Engineering, and several technical roles at eBay. Sid earned his BS and MS degrees in CS from Cornell University, where he focused on Distributed Systems. Outside of work, Sid is a maintainer/committer on Apache Airflow and advises early-stage companies and several conferences (QCon, Data Council, and conferences under Skills Matter).

Detecting Deception and Tackling Insurance Fraud Using Conversational AI

We are developing an explainable pipeline that will identify and justify the behavioural elements of a fraudulent claim during a telephone report of an insured loss. To detect the behavioural features of speech for deception detection, we have curated a robust set of acoustic and linguistic markers that potentially indicates deception in a conversation. Using statistical measures and machine learning approaches, the detection of these linguistic markers in the right context is being investigated. The explainable pipeline means that the output of the decision-making element of the system will provide transparent decision explainability, overcoming the “black-box” challenge of traditional AI systems.

Dr Julie Wall is a Reader in Computer Science, Director of Impact and Innovation for the School of Architecture, Computing and Engineering and leads the Intelligent Systems Research Group at the University of East London. Her current research focuses on developing machine learning and deep learning approaches for speech enhancement, natural language processing and natural language understanding and she maintains collaborative R&D links with industry. This has led to the successful acceptance of two Innovate UK grants with a combined total value of £2,273,177. Since starting her PhD in 2006, Julie has been exploring the overarching research area of designing intelligent systems for processing and modelling temporal data. This primarily involves investigating the architectures and learning algorithms of neural networks for a variety of data sources.

https://www.uel.ac.uk/research/intelligent-systems

Low Latency Fraud Prevention with Redis Labs

Time is money, and the difference between milliseconds and microseconds can mean the difference between big profits or losses. In modern business low latency is important, but it’s critical in banking and financial services, especially since fraud has grown to epidemic proportions. Fraudulent charges are costing companies billions of dollars, and that’s why it’s important for IT decision makers in financial services to measure latency in microseconds. A low-latency, scalable, in-memory data store like Redis can enable organizations to architect to meet this critical business need. Drawing from first principles and best practices, learn how to architect to prevent fraud in milliseconds.

Key Takeaways:

*Understand the importance of fraud prevention, especially given the rise of digitalization

*Become familiar with first principles of real-time data fraud prevention, and its associated platform components

*Appreciate the technical merits of in-memory databases such as Redis in facilitating this architecture

*Learn best practices of implementing low latency fraud prevention based on examples from the industry related to 1) e-mail fraud and 2) payment fraud

As Chief Business Development Officer, Taimur is responsible for developing emerging businesses at Redis Labs, and strategic business & corporate development. He is currently leading initiatives related to AI/ML. Prior to Redis Labs, Taimur led Worldwide Customer Success for Microsoft’s Azure Data Platform, Analytics & AI business. And before that, he was Managing Director for Amazon Web Services (AWS) Platform Technology and Applications where he led business development from 2008 (near its inception) to 2018 when the business reached $25B in ARR. Taimur helped forge some key partnerships and customers including Airbnb, CapitalOne, Dropbox, Liberty Mutual, NASA JPL, Nasdaq, Netflix, Nintendo, Intuit, SAP, Samsung, and Societe Generale.

How to Succeed with Machine Learning Against Fraud

Applying data science to predict fraud becomes an industry standard. But machine learning in fraud has its own specifics compared to other domains: highly imbalanced data, different cost of false positives and false negatives, interaction of ML models with the whole fraud system, etc. In my talk, I will share what to do with such caveats: how to choose the correct metric for model training based on business problems, how to improve models with undersampling methods, what to do in case of real-time fraud detection systems. Also, I will talk about an important business part: how to measure impact of ML models and why sometimes you need to let fraudsters in.

Elizaveta is a Senior Data Scientist at Bolt, leading European mobility platform. Her current focus is on commerce topics where she applies models and algorithms to reduce fraud and optimise payments; she is involved in the whole data science pipeline starting from problem definition and model prototyping to live ab-tests and model operations. Being passionate about math and having degrees in finance and quantitative economics, Elizaveta transitioned to Data Science from Business Analytics, marking her path with numerous math competitions and hackathons. Elizaveta is active in Data Science community in Estonia (organising and speaking at conferences and meetups). She is passionate not only about Data Science, but also about engaging more people into the tech industry (organizing Tallinn Django Girls workshop, speaking at events and camps for school kids about career in IT).

Financial Crime & Fraud Detection Using AI: Focus in Graphs: Key applications, Lessons Learned & Industry Trends

In recent years and specially during the COVID 19 pandemic, there has been an unprecedented growth in digital ecommerce that has significantly increased payments. This in turn has increased considerably fraud and financial crimes. In this recent landscape, traditional fraud detection approaches mainly based in rule-based engines presented a number of growing challenges, which both an ever growing number of cases but more importantly the number of false positives being significantly increased which has put a strain in the Fraud and Financial crime investigation teams. In this context, AI and in particular ML approaches using graphs has gained a lot of traction in the marketplace. Whilst these solutions have indeed significant potential to address the challenges of the legacy rule-based system, there are also a new type of considerations that must be observed. We will explore during the session some use cases and how to successfully addressed the main challenges implementing these graph systems. We will also explore future trends

Javier Campos is the Head of the Experian DataLabs for UK&I and EMEA. He oversees innovation with emphasis on the development of new Data-driven products and services. Pioneering application of Artificial Intelligence in mobile, voice, fraud, credit, marketing, social media, digital advertising and healthcare. Javier is also a permanent member of the Bank of England & FCA Artificial Intelligence Public-Private Forum, an industry group created to accelerate AI adoption within the Finance industry. He is a frequent speaker at industry conferences. Javier has 28+ years’ experience globally within the Finance, AI, Market Research, Media and Technology fields including serving at Global Chief Technology Officer at Kantar-WPP, Havas Media and CIO at GroupM-WPP.

Krishna Sankar is a Distinguished Engineer – AI, at the Enterprise Analytics & AI (EAA) of U.S. Bank, focusing on embedding intelligence in Financial Systems; all aspects incl the E3 of AI - Explainability, Experimentation & AI Ethics. Earlier stints include Senior Data Scientist/Volvo, Chief Data Scientist/blackarrow.tv, Data Scientist/Tata America Intl, Director of Data Science/Bioinformatics startup & as a Distinguished Engineer/Cisco. His external work includes teaching, writing blogs, researching chess openings and Lego Robotics. He has been speaking at various conferences incl NeurIPS 2020 (Moderated 2 panels), RE.WORK [bit.ly/3712jKG], Nvidia GTC2019 and GTC2020, ML tutorials at Strata SJC & London, Spark Summit and others. His occasional blogs can be found at https://ksankar.medium.com/ viz. Conversational AI [bit.ly/37PjPBh], Rebooting AI [bit.ly/2ststqi41], The Excessions of xAI [bit.ly/2LDXe2c], Robots Rules[goo.gl/5yyRv6], NeurIPS 2018 [goo.gl/VgeyDT], Garry Kasparov’s Deep Thinking [goo.gl/9qv671] and a lot of Sci-Fi and crime noirs ! He has been guest lecturing at the Naval Post Graduate School, Monterey; and you will find him occasionally at the FLL (Lego Robotics) World Competition as Robots Design Judge.

Mairtin is the CIO at Ravelin, which provides fraud protection for online businesses. The “I” in CIO stands for “Intelligence”, and Mairtin runs the Data Science and Investigations team in Ravelin, responsible for the machine learning and graph network models at the heart of Ravelin’s detection engine. He was previously the Head of Fraud at the taxi app Hailo, where he honed his fraudster frustrating skills. He is an internationally experienced intelligence analyst, with stints in Scotland Yard, the United Nations and elsewhere.

Mousumi is a Risk Analytics Leader from San Francisco, CA with a background in Systems Engineering with specialization in Data Science who focuses on solving critical business problems for top Consulting, Fintech and Financial Services companies in Payment Fraud and Money Laundering domains through Data Analytics, and, help banking organizations in saving multi million dollar fraud and operational loss as well as in meeting regulatory guidelines. She earned a Masters in Systems Engineering from Virginia Tech University and has Data Analytics/Data Science certifications from Columbia, Stanford, and San Diego University. Mousumi has technical expertise in Data Analytics, Data Quality, Data Science, Data Visualizations, Text Analytics, Payment Fraud/Money Laundering, Analytical Databases, SQL, SAS, Python, and Tableau/Power BI.

Satheesh is currently leading a global seller risk data science team at eBay, which leverages cutting edge AI/ML techniques to monitor and manage risk across the seller life-cycle within eBay. Prior to this, he was the Principal & Co-Founder of Elite Analytics LLC, which was acquired by Aviana Global in 2017, where he served as the director of data science solutions. His 20+ years of experience spans multiple domains, including fraud and compliance, healthcare, media, retail, financial services, e-commerce, customer analytics, and travel & hospitality. He has specifically worked with AI/ML based fraud and compliance detection in the areas of insurance, warranty, tax, ecommerce, cyber security and others. He has a M.S. and Ph.D. in statistics and Operations research from Texas A&M university, and an undergrad from Indian Institute of Technology Kharagpur.