Crimtan Announces Exclusive Data Science Event in Partnership with Cardiff University
24 Apr, 2017
On June 8th Cardiff University and Crimtan will host an exclusive Data Science event:
LOOKING INSIDE THE BLACK BOX. HOW DATA SCIENCE CAN BE APPLIED TO IMPROVE DIGITAL ADVERTISING RESULTS.
Taking place at The Royal Institution of Great Britain, in the heart of London’s Mayfair, the event will run from 2pm to 6pm and will be followed by a drinks reception.
For the last two years, Crimtan and the Department of Mathematics at Cardiff University have been working together to gain a deeper understanding of how data science can improve digital advertising performance. During the afternoon, four presentations by Professors at the University and Crimtan will reveal the very latest research into programmatic advertising, and show how advertisers can implement new data strategies to drive better ROI from their digital marketing budget. The event will close with a question and answer session featuring a panel of experts.
Admission is free, but to reserve your place, please contact firstname.lastname@example.org.
Places are limited, so tickets will be allocated on a first come, first served basis.
Presentations will cover the following topics:
Lifting the lid – Inside Programmatic
This presentation will look behind the recent boost in data-driven, programmatic advertising and examine the size of data sets involved, the challenges and benefits to marketers along with the latest developments in statistics, data science and ad tech. It will assess the current state of decision making and, using examples from inside and outside digital advertising, explain why algorithms, AI and machine learning are at the heart of it.
Research Report One
Artificial intelligence and machine learning for devising bidding strategies
In this talk, we give a brief overview of the statistical and machine learning techniques that have been used in analysing online advertisement data and devising programmatic bidding strategies. The machine learning techniques examined include support vector, gradient boosting and factorization machines and along with some basic techniques of extracting information about dependencies between behavioural characteristics of the users and their conversions. As well as showing the results of a statistical analysis of large data sets, this presentation will compare several novel decision making approaches and reveal methods of extracting information from data sets and demonstrate how they make a difference to performance.
Research Report Two
Models for evaluating advertisement efficiency.
How an advertisement can impact on user choice, change in preference and brand loyalty. In this talk we describe a simple model for evaluating the efficiency of an advertisement. This model assumes that after being exposed to an ad, potential buyers can change their preference either at random or after being influenced by the ad. This model gives more realistic estimates of the ad efficiency than the simpler model which ignores random change of preferences. We illustrate the discussion by showing results of data analysis and demonstrate that the probability of randomly changing preferences differs a lot across categories.
What it means for you
Summing up the research findings, we will show how the latest statistical techniques and technical developments can help advertisers reach relevant audiences and engage them with personalised messages that get a better response and greater return on investment from digital marketing budgets.
ABOUT THE PRESENTERS
Anatoly Zhigljavsky PhD
Anatoly graduated from the Faculty of Mathematics, St. Petersburg State University, in 1976 and gained his PhD on applied probability in 1981. After holding the post of Professor of Statistics at the St. Petersburg State University from 1989-1997, Anatoly became Professor, Chair in Statistics at Cardiff University in 1997. He has been the author or co-author of nine monographs on the topics of stochastic global optimization, time series analysis, optimal experimental design and dynamical systems; the editor/co-editor of 8 books on various topics, and the author of about 150 research papers in refereed journals. In addition, Anatoly has organised several major conferences on applied statistics, time series analysis, experimental design and global optimisation. A member of the editorial board of two journals: Journal of Global Optimization and Statistics and Its Interface, Anatoly has been involved with applied projects on marketing research and consumer behaviour for more than twenty years and has worked with many major corporations, including Procter & Gamble, ACNielsen and GlaxoSmithKline.
Andrey Pepelyshev PhD
Andrey Pepelyshev graduated from the Faculty of Mathematics at St. Petersburg State University, Russia in 2000 and obtained his PhD on applied probability in 2002. He lectured at St. Petersburg State University from 2003-2007 and was a research fellow at Venice University in 2008, at Sheffield University in 2009-2010, and at the RWTH Aachen University in 2011-2013. Since 2013, he is a lecturer at the School of Mathematics, Cardiff University and has about 40 papers published in leading statistical journals. Andrey’s research interests are concentrated around the theory of optimal experimental design, analysis and modelling of experimental data, and analysis of complex models and big data. He was actively involved in a project between Cardiff University and Procter & Gamble in the field of statistical modelling in market research and also participated in a project between Cardiff University and GlaxoSmithKline on clinical trial studies.
Nina Golyandina PhD
Graduating from the Faculty of Mathematics & Mechanics, St. Petersburg State University, in 1985, Nina gained her PhD on applied probability in 1998. An Associate Professor at the St. Petersburg State University since 2000, she has co-authored 2 monographs on time series analysis and is author of about 30 research papers in refereed journals. Nina has been involved with applied projects on marketing research and consumer behaviour for more than twenty years.
Yuri Staroselskiy PhD
Chief Technology Officer Crimtan
Yuri started his career in advertising in 2007, joining the research team at Nebuad. Yuri worked with big data, using his academic expertise in experimental design to develop new algorithms. Two years later Yuri joined Red Aril as head of ad operations, helping the company build a team of traffickers and analysts. Yuri formally took the role of Head of Technology in March, 2013. Yuri holds a PhD in Statistics from Saint Petersburg State University
Rob Webster BSc Mathematics
Chief Strategy Officer Crimtan
Rob has over 16 years of ad tech expertise gained from working with some of the world’s largest advertisers, most recently as Data and Technology Director at MediaCom. Rob is a regular speaker on the conference circuit and contributor to the advertising media. Prior to MediaCom Webster’s career included senior roles in data and ad tech at Unique Digital, Tacoda and Yahoo.
Thursday, June 8th
2 pm to 6 pm (with drinks reception to follow)
The Conversation Room
The Royal Institute of Great Britain
21 Albemarle Street
London W1S 4BS