Past Meetup

Audience Science: User Modelling and Behaviour Targeting

This Meetup is past

60 people went

Room A1/3, 3rd floor, Physics Building, UCL

Gower St, WC1E 6BT · London

How to find us

The best way to find the room in Physics Building is through main quad of UCL in Gower Street, which is near the Euston Square tube station. We will have people greeting you at the main gate of UCL on Gower Street.

Location image of event venue


The topic of the second event of London Computational Advertising & Behaviour Targeting Group will be the user modelling & behaviour targeting and their applications in online advertising.

Behavioral targeting comprises a range of technologies and techniques used by online website publishers and advertisers aimed at increasing the effectiveness of advertising using user web-browsing behaviour information to select advertisements to display.

Behavioral marketing can be used on its own or in conjunction with other forms of targeting based on factors like geography, demographics or contextual web page content. It's worth noting that many practitioners also refer to this process as "Audience Targeting".

We’re delighted to have two very distinguished speakers from both industry and academia to talk about their research and practice of user modelling and behaviour targeting, their technical challenges and research problems.

They are:

Natasa Milic-Frayling, Principal Researcher from Microsoft Research

Topic: The Web’s Invisible Net

Natasa Milic-Frayling will be talking about: “The Web’s Invisible Net”, shedding light on the user tracking and behavioural targeting practices. Through empirical studies of cookie tracking practices, Natasa will provide an alternative view of the display ad business by observing the network of third party trackers that envelopes the Web. The practice begs a question of how to resolve the dissonance between the current consumer tracking practices and the vendors’ desire for consumer’s loyalty and trustful long-term engagement.

Speaker Bio

As a Principal Researcher at Microsoft Research (MSR) in Cambridge, Natasa is working on the design, prototyping and evaluation of information and communication systems. She is passionate about innovation in personal and social computing and promotes a dialogue between IT industry, consumers, and policy makers on the issues that arise from the adoption and use of technology. Her current focus is on digital obsolescence, activity based computing across devices and services, and privacy respecting systems and applications. Natasa is actively involved with a wider community of academics and practitioners through public speaking, collaborative projects, and serving on advisory boards of academic programs and commercial enterprises. She is a Visiting Professor at the UCL and QMUL and a member of the ACM EU Council.

Iason Bakogiannis, Engineering Director - Infrastructure at Affectv

Topic:Practical challenges of enrichment of bid request data

Affectv will demonstrate the steps and technologies used in tackling the problem of enriching bid-request data with additional information. The talk will focus on device type identification and the process of making the data available to bidding agents in real-time.

Company Info

Affectv's mission is to make advertising more relevant for people so it can help improve the Internet. Affectv helps brands and marketers to find, understand and connect with the relevant audience, in real-time across multiple digital channels, by translating the world of noisy data into peoples' signals.

website (

Team Bio

Iason Bakogiannis has been with Affectv for almost 3 years, he has a Masters in Intelligent Systems from KCL. In the past he has worked on big data machine learning applications and is currently heading the development of real-time systems and applications at Affectv as Infrastructure Director.

Austin Leirvik has worked as a data scientist at Affectv for the past two years. His work at Affectv focuses on integrating different data sources to improve bid and campaign delivery optimization. He also has a background in natural language processing and is heavily involved in developing new approaches to web-scale text classification.

Miguel Sempere is a real-time systems developer, passionate about all that can be compiled. Before Affectv, he worked as a Embedded engineer for a social robots company, where bring things to life was his duty.

Konstantinos Bozas recently finished his PhD in Computer Vision and Machine Learning from Queen Mary University of London with specialization in sketch based image retrieval and sketch recognition. He is currently Data Scientist at Affectv.


18h00 - 18h15: Shuai gives a welcome talk, introduce the group, guests and thank sponsors and state the rest of the proceedings.

18h15 - 19h00: Natasa Milic-Frayling from Microsoft Research

19h00 - 19h30: 30 min break - beer and pizza served

19h30 - 20h15: Iason Bakogiannis (and this team) from Affectv

20h15 +: Drinks and networking.