DSPT is back for another great year! This time we begin in Lisbon learning about what takes to bring data science to a usable product.
Pedro Balage from Farfetch will talk to us about the new advances in graph representations and how they can be useful for ECommerce companies such as Farfetch.
João Veiga from Feedzai will, in turn, provide us the perspective of a Data Scientist and the challenges it faces daily within a company.
Did we spark your interest?
Come and join us!
=== SCHEDULE ===
The preliminary agenda for the meetup is the following:
• 18:30-19:00: Welcome and get together
• 19:00-19:30: Talk 1: "New advances in Graph Representations for Ecommerce Search" by Pedro Balage - Farfetch
• 19:40-19:45: Group photo
• 19:45-20:15: Networking / Coffee Break
• 20:15-20:45: Talk 2: "3 Shades of Data Science" by João Veiga - Feedzai
• 20:50: Closing, hanging out and some beers
• 21:00: Dinner is optional but it might be an excellent opportunity for networking
If you want to attend the dinner please check in here: https://doodle.com/poll/mkmgwdyii6upc9v5
We would like to thank Feedzai for its partnership and support in this event!
Do you want to be a sponsor in future meetups? Please contact us to [masked]
Talk 1: New advances in Graph Representations for Ecommerce Search - Pedro Balage
Abstract : This talk explores new advances in deep neural networks and graph representations using click-through logs to improve ecommerce search. It focuses on providing query-document relevance by modeling click-through logs with graph embeddings and other related techniques. The application of such methods goes beyond information retrieval as it is also applicable to areas such as recommendation systems, anomaly detection, clustering, among others. This talk will provide a nice overview of the recent literature from top-tier conferences and will show the basic concepts of what is being considered a new trend in machine learning and data science.
Short Bio: Pedro Balage works as Lead Data Scientist at Farfetch, focusing on providing semantic capabilities for its ecommerce search. He holds a PhD in Natural Language Processing from the University of São Paulo (Brazil) and he is affiliated to the Instituto de Telecomunicações (IT - Lisbon). He is member of the organization committee for the Lisbon Machine Learning School (LxMLS).
Talk 2: 3 Shades of Data Science - João Veiga
Abstract : What drives you the most? Getting through that pesky target right before the submission deadline? Implementing and testing that fancy but obscure algorithm you saw on KDD? Maybe it's all of that but in moderate dosage.
This is where the Product Data Scientist comes in. Through cross-functional teams, clear goals, and with lots of communication we are able to focus on solving our customer's pain: fraud.
Join me in uncovering the different shades of data science in a "data-heavy" company whose products are built and used by data scientists to fight fraud in real time around the globe.
Short Bio : Since joining Feedzai on 2015, after having worked on a Public Hospital to successfully help improving organ donation, João has been through the trials and tribulations of using Machine Learning to help making commerce safer.
Protecting consumers and retailers from all sorts of fraud schemes, in real time, across >100 countries, in highly unbalanced situations, is no small feat. But for this Biomedical Engineer turned fraud fighter, his bread and [vegan] butter is solving tough real world problems with just the right mix of engineering and data science.