Departmental Seminar: Media Futures - Emine Yilmaz, Jun Wang and Shi Zhou

Speaker: Emine Yilmaz, Jun Wang and Shi Zhou
UCL Contact: Sarah Turnbull (Visitors from outside UCL please email in advance).
Date/Time: 18 May 16, 12:00 - 14:00
Venue: 1.03

Abstract

Task based Information Retrieval by Emine Yilmaz

Abstract: The need for search often arises from a person’s need to achieve a goal, or a task such as booking travels, organizing a wedding, buying a house, investing in the stock market, etc. Since current search engines focus on retrieving documents relevant to the query submitted as opposed to understanding and supporting the underlying information needs (or tasks) that have led a person to submit the query, search engine users often have to submit multiple queries to achieve a single information need. For example, booking travels to a location such as London would require the user to submit various different queries such as flights to London, hotels in London, points of interest around London as all of these queries are related to possible subtasks the user might have to perform in order to arrange their travels.Ideally, a search engine should be able to understand the reason that caused the user to submit a query (i.e., the actual task that caused the query to be issued), and rather than just showing results relevant to the query submitted, it should help the user achieve the actual task by guiding her through the steps (or subtasks) that need to be completed. Devising task based information retrieval systems have been the recent focus of my research group. In this talk, I will talk about the progress we have made in this direction.

Bidding machines in real-time advertising by Jun Wang

Abstract: In display advertising, impressions are aggregated and traded in real time in ad exchanges, and their transaction volume has already surpassed that of the financial market. In this talk, I will present our work on "learning to bid", which allow advertisers to bid on a display ad impression in real time when it is being generated. It goes beyond contextual advertising by motivating the bidding focused on user data and it is different from the sponsored search auction where the bid price is associated with keywords. Our solutions have been successfully deployed and tested in various large-scale operational advertising systems.

Critically hybrid epidemic spreading — from Internet worm Conficker to HIV in vivo infection by Shi Zhou

Abstract: Many epidemics in nature, cyberspace and society use multiple spreading mechanisms simultaneously. We showed that a combination of simple and individually ineffective spreading mechanisms could cause massive outbreaks. Our study explained why the Internet worm Conficker was able to infect millions of computers in just a few days in 2008 and is still active today. In collaboration with immunologists at UCL and Oxford, we proposed the first model that reproduced the full course of HIV in vivo infection and progression to AIDS.