DSPT#57 - Handling missing data and selling the right thing (Coimbra)


Have you ever wondered on how to deal with the common and significant problem of missing data?

Or are you more interested on how Altice Labs manage to sell the right thing to the right person at the perfect moment?

Join us in Coimbra to gain a better understanding of the aforementioned topics.

This meetup is sponsored by Talkdesk (https://www.talkdesk.com). 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: Handling Missing Data with Imputation - Ricardo Pereira
• 19:40-19:45: Group photo
• 19:45-20:15: Networking / Coffee Break - by Talkdesk
• 20:15-20:45: How ML generates new insights to create Contextual marketing - Ricardo Filipe
• 20:50: Closing
• 21:00: Dinner is optional but it might be an excellent opportunity for networking. (register here: https://doodle.com/poll/ye725h88tddedynn )

Talk 1
Title: Handling Missing Data with Imputation

Abstract: Missing data can be defined as the lack of information in one or several features of a dataset and it compromises the performance of most classifiers. This issue is usually handled with data imputation, where new plausible values are generated to replace the missing ones. In this talk a brief introduction to the missing data concepts is going to be presented, followed by a discussion about the state-of-the-art methods used to perform imputation.

Short bio: Ricardo Cardoso Pereira is an Invited Assistant teacher and PhD student at the University of Coimbra, where he also got his master’s degree in Informatics Engineering. He also teaches at the Coimbra Institute of Engineering, where he got his bachelor’s degree. Ricardo is currently affiliated with the Centre for Informatics and Systems of the University of Coimbra (CISUC) and in the past years he has worked as a Research Fellow on the XENON Dark Matter Direct Search Experiment. His main interests are computational intelligence and data processing, with special emphasis on the missing data field.

Talk 2:
Title: How ML generates new insights to create Contextual marketing

Abstract: Telecommunication operators are evolving to retrieve new services, improve customer’ quality-of-service and increase operational efficiency. Machine Learning techniques have an important role to create new tools or improve current Altice Labs products. More specifically, in this presentation, we are going to focus how machine learning generates new insights to create Contextual marketing. Contextual marketing, also known as Behavioral Targeting, is an online and real-time marketing model that allows companies to sell the right thing to the right person at the best moment and increase their customer base. Interpretable machine learning models can improve the way how operators analyze the aforementioned marketing campaigns. Additionally, it will allow to analyze with little effort the patterns of clients reducing the time needed by business analysis teams.

Short bio: Ricardo Ângelo Filipe, completed his master's degree in Electronic Engineering and Telecommunications in 2008, having joined the former PT Inovação (now called Altice Labs). He has been involved in several technological upgrading projects in different telecommunication operators within different geographical areas. More recently, he has been involved in the introduction of machine learning in some of Altice Labs' products. Additionally, he is a PhD student at the University of Coimbra, on the field of monitoring of distributed systems.