Скачать с ютуб Prototyping with R Packages: Case Study on E-Commerce Fraud Detection - Irene Steves & Yogev Herz в хорошем качестве

Prototyping with R Packages: Case Study on E-Commerce Fraud Detection - Irene Steves & Yogev Herz

In their work, Irene and Yogev often come across fraud patterns in eCommerce that their regular models cannot detect. While retraining models on newer data can solve some of these problems, some patterns require entirely new analyses. In this talk, Irene and Yogev discuss how they tackled fraudsters taking advantage of "Buy Online Pay in Store" policies, and how R packages helped gradually transition from analyses to production-ready code. Irene holds an M.Sc. in Ecology and a B.A. in Integrative Biology, through which she first discovered R and data science. Her interest in data-led her to the Arctic Data Center at the University of California Santa Barbara, a summer internship at RStudio, and ultimately to the Research department at Riskified, where she now explores the fascinating patterns of fraud in e-commerce. Yogev first encountered R when studying the social organization of bumblebees for his M.Sc. in evolutionary biology. He transitioned to full-time R user and programmer when he left the hot and humid beehive to the air-conditioned office space of the research department in Riskified. He enjoys learning new R tricks and packages to improve the department’s data science capabilities. The talk was part of R-Ladies Amsterdam community event -https://www.meetup.com/rladies-amsterdam For more meetups, join our Meetup group - https://www.meetup.com/Meetups-at-Ris... And check our Tech Blog - http://riskified.technology/