DataFramed #65 Preventing Fraud in eCommerce with Data Science
Jun 28, 2021
Elad Cohen, VP of Data Science and Research at Riskified, shares insights on leveraging data science for fraud detection in eCommerce. He discusses his fascinating transition from physics to data, emphasizing the impact of machine learning in real-time fraud prevention. The conversation covers best practices for building effective detection systems and the vital collaboration between data scientists and engineers. Elad highlights the importance of data-driven decisions and staying updated with new technologies to foster success in this rapidly evolving field.
AI Snips
Chapters
Transcript
Episode notes
Elad's Career Journey
- Elad Cohen's background started in physics, researching lasers, then transitioned to signal processing in the military.
- Realizing a love for data and research, he shifted to data science in 2013, utilizing online courses like DataCamp.
E-commerce Acceleration
- The COVID-19 pandemic accelerated e-commerce adoption by several years, forcing businesses to digitize and consumers to embrace online shopping.
- This shift is permanent as consumers now recognize the significant advantages of online shopping experience.
Merchant Fraud Liability
- Merchants bear liability for fraudulent online credit card transactions, unlike in-person purchases where card companies are liable.
- This risk incentivizes merchants to be conservative, sometimes blacklisting entire countries or using inflexible rule-based systems.