DSPT#36 (Porto) - Machine Learning in Retail and Fraud Detection
Details
DSPT is back from the holidays and this time we will talk about the center of each business: customers. Miguel Arantes from Inovretail will talk to us about how Inovretail applies Machine Learning to one of the most aggressive markets: retail. Can machine learning help retailers ? We’ll see…
Besides getting and keeping customers, in today’s internet age, we also need to protect them and ourselves and Feedzai is on the forefront of fraud detection. But how can Feedzai stay ahead of the curve and fight fraudsters? Miguel Araújo will unveil a little bit of the research happening at the company and how data science helps them stay one step ahead of the criminals of the XXI century.
Curious?
=== SCHEDULE ===
The preliminary agenda for the meetup is the following:
• 18:30-19:00: Welcome and get together
• 19:00-19:30: Talk 1: Customer retention, the machine learning strategy by Miguel Arantes, Inovretail.
• 19:40-19:45: Group photo
• 19:45-20:15: Networking / Coffee Break
• 20:15-20:45: Talk 2: The Forefront of Fraud Fighting with Research @ Feedzai by Miguel Araújo, Feedzai
• 20:50: Closing, hanging out and some beers
• 21:00: Dinner is optional but it might be an excellent opportunity for networking. Please register at: https://doodle.com/poll/iphe9whcpsh2gq9r
This meetup is supported by Feedzai (https://feedzai.com/) and DevScope (https://www.devscope.net/).
Do you want to be a sponsor in future meetups? Please contact us to info@datascienceportugal.com
See you there!
-----------------------------------
Talk 1: Customer retention, the machine learning strategy.
Abstract: Inovretail provides data science solutions and digital tools, for retail, that deliver quantifiable insights and actionable recommendations with direct and sustainable impact on key metrics.Our solutions are SaaS based, fast to deploy, easy to use and help retailers in three key areas:Stock Recalibration, Digital Sales Assistant and Multi-dimensional Forecasting.Within this context, Data Science plays a key role in the company: making the right questions and providing the best answers based on data. In this presentation, I will talk about one of our solutions to help the retailers. The Churn Analytics, our Data Science solution that returns a list of potential churners, enabling you to pro-actively develop churn mitigation strategies.
Short bio: Miguel Arantes holds a master in Mathematics Engineering from the University of Porto. He has been working in statistical models for over 8 years in credit institutions and IT companies. His priority is his family (2 loving kids), plays Pokemon GO, these days learning MongoDB, using the lintR library and also doing a few articles for fun in LinkedIn.
Talk 2: The Forefront of Fraud Fighting with Research @ Feedzai
Abstract: Prepare for a ride along the fraud-fighting pipeline and multiple research initiatives at Feedzai. We'll see a) how fraud starts and how transaction records can be used to identify databreaches and other points-of-compromise; b) how features based on historical records can be computed (actually, estimated!!) in very low latency environments; c) how we leverage Deep Learning to enhance the predictive performance of our product d) how our data scientists' work is facilitated with our AutoML efforts and; e) how we deal with the reject inference problem and avoid poisoning our training datasets.
Short bio: Miguel Araújo is a senior data scientist at Feedzai. Before joining the fraud-fighting cause, Miguel obtained his PhD in Computer Science at Carnegie Mellon University and the University Porto where he developed anomaly detection, pattern mining and prediction solutions in various domains using telecommunication records and social network graphs.