Logo Crossweb

Log in

No account yet? Forgot password

Przypomnij hasło

close Wypełnij formularz.
Na Twój adres e-mail zostanie wysłane link umożliwiający zmianę hasła.
Send
This event has already taken place. Check upcoming events

Data Science Meetup #45

Event:
Data Science Meetup #45
Event type:
Meetup
Category:
IT
Date:
14.05.2019 (tuesday)
Time:
18:00
Language:
English
Price:
Free
City:
Place:
Sala 107 Wydział MINI PW
Address:
Koszykowa 75
Description:

The next Data Science Warsaw Meetup will take place on Tuesday, 14th May, and afterward we invite you to the informal Networking powered by NETHONE! During the meetup we'll have two talks:


1. „Data Science in RTB: Data Pipeline from Messy Data to Shiny Coins” Radosław Kita


Real-Time Bidding is a fascinating world like from James Bond movies. Everything takes place between clicking on the link and displaying the page. We have superfast servers, beautiful data and machinations.


I would like to talk about our world. About our machine learning pipeline: the way from data to models.

Radosław Kita Team Leader @Adform.

He was responsible for creating analytical or reporting systems in Onet, Alior Bank, TVN, Allegro, Veon (sixth telecom in terms of number of users in the world), Showmax. Currently, he manages a team DSP Algos at the AI & Research department.

He participated in projects to build from the scratch DMP systems, integrating data on customer behavior from internal systems, social media and external systems, or a platform for conducting A/B tests. He dealt with churn modeling, customer value in time, customers clustering and reconstructing demographic characteristics based on behavior.


2. „RTB Bid Landscape - survival models powered by bitmaps” Łukasz Mączewski, Przemek Piotrowski


The Bid Landscape (BL) is one of the key components of the bidding strategy optimisation in the area of the Real-Time Bidding (RTB). It allows for predicting RTB campaign performance with respect to its targeting and budget settings. The essential element of BL is the ability for predicting auction winning probability for the specific bid price. During presentation, we will go through technological and data science challenges regarding huge data streams we are facing (millions of auctions per second) and right-censored data we are to model (only small fraction of all winning prices is known, for the remaining ones only lower win price boundary is available).


Lukasz Maczewski Senior Data Scientist in DSP Algos team in Research Area. He is an ex-physicist, holding PhD in an experimental, particle physics and a senior data scientist in the Adform research team. In his current role, he is responsible for developing, deploying and maintaining real-time bidding (RTB) algorithms for online advertising. In his former job he was supporting analytics-based maintenance for General Electric turbofan jet engines. His specialization were advanced reliability models and diagnostics algorithms for failure detection. Łukasz has over 10 years of experience in research and widely understood data science.

Przemek Piotrowski Senior Data Engineer@ Adform, AI& Analytics. Responsible for Bid Landscape, Forecasting and Bidding Algorithms pipeline. Contributor to Roaring Bitmaps JVM implementation.


After the meetup we invite you to an informal Networking powered by Nethone. Before the meeting, send questions and problems regarding data science, machine learning or the fight against fraud that you have encountered. Specialists from Nethone will suggest appropriate solutions - they will also indicate the best practices used in working with data. More about Nethone http://www.nethone.com


We also encourage you to subscribe to the newsletter (https://tinyletter.com/datascience), where we inform about interesting events regarding Data Science in Warsaw and in Poland.

Profile of employers

Similar events