DSPT#59 - May I recommend some Popcorns of knowledge? (Porto)
Details
One of the pains of 21st century life is telemarketing and customer service calls....can we make them less intrusive and actually useful for the customers? We might have someone who knows how… no need for a google search. Nobody goes beyond the first page anyway, wanna know why? Come hear about it! Come for the popcorns, stay for the knowledge.
This meetup is sponsored by NOS (http://www.nos.pt/ ) and COCUS (https://www.cocus.pt). Cheers and thank you for the support!
=== SCHEDULE ===
The preliminary agenda for the meetup is the following:
• 18:30-19:00: Welcome and get together
• 19:00-19:30: Best Time for Contact by João Rafael
• 19:40-19:45: Group photo
• 19:45-20:15: Networking / Coffee Break
• 20:15-20:45: Ranking: Why you never go past the first page on a search engine by Pedro Nogueira
• 20:50: Closing
• 21:00: Dinner is optional but it might be an excellent opportunity for networking (register here: http://bit.ly/dinner_dspt59)
Talk 1
Title: Best Time for Contact
Abstract: As part of its daily activities, NOS needs to be able to call its clients to communicate discounts, contractual terms, resolve technical issues, etc. Knowing when to contact each client ensures that inconvenient calls are avoided and improves call center operations. The Best Time for Contact project showcases:
- How to use ML to predict which calls will be answered;
- How to combine these individual predictions to find the best calling strategy using linear programming and heuristics;
- A troubled road to production: legacy data issues, integrations, war-stories and much more.
Short bio: João Rafael is a MSc in Software Engineering from University of Coimbra. For 2 years he conducted research on High Performance Computing at CISUC. Afterwards, he worked at Feedzai for 4 years where he improved the product's Machine Learning internals and worked worldwide with several clients as Solutions Architect. He founded is own Data Science consulting company Upper Delta and is working together with NOS to improve their DS initiatives and products.
Talk 2:
Title: Ranking: Why you never go past the first page on a search engine
Abstract: One can say that designing a ranking engine is a mixture of art, science and engineering and be very right about it. This presentation will do a small introduction of the problems that come up with building a ranking engine and the tools that can be used to solve them.
Short bio: Pedro Nogueira is a Data Scientist at Farfetch, working on Information Retrieval. He holds an MSc in Electrical and Computer Engineering from FEUP, where he first dabbled in Data Science using Matlab (never again). He thoroughly enjoys designing scalable and robust machine learning applications while making sure that all the best data science practices are being used. He also used to play professional volleyball but no longer knows what a ball is.
