[PhD student talk] Online Learning in Multi Period Information Retrieval

[ October 24, 2012; 4:00 pm to 5:00 pm. ] Location: Roberts, 110 Speaker: Marc Sloan Title: Online Learning in Multi Period Information Retrieval Abstract: In Multi Period Information Retrieval we consider retrieval as a stochastic yet controllable process, the ranking action during the process continuously controls the retrieval system’s dynamics, and an optimal ranking policy is found in order to maximise the overall users’ satisfaction. Our derivations show interesting properties [...]

[PhD student talk] CIKM rehearsal – Sequential Selection of Correlated Ads by POMDPs

[ September 26, 2012; 4:00 pm to 5:00 pm. ] Location: MPEB 6.12a Speaker: Shuai Yuan Title: Sequential Selection of Correlated Ads by POMDPs Abstract: Online advertising has become a key source of revenue for both web search engines and online publishers. For them, the ability of allocating \emph{right} ads to \emph{right} webpages is critical because any mismatched ads would not only harm web users’ satisfactions but also lower the [...]

[Invited talk] Understanding Querying Behavior through Simulation

[ September 26, 2012; 11:00 am to 12:00 pm. ] Location: Anatomy B15 Title: “Understanding Querying Behavior through Simulation” Speaker: Dr. Leif Azzopardi, Researcher of Information Access and Retrieval,  Lecturer of Web Application Development, School of Computing Science, University of Glasgow. Abstract: Typically, Information Retrieval evaluation focuses on measuring the performance of the system’s ability at retrieving relevant information, and not the query’s ability. However, the effectiveness of a retrieval system [...]

[Invited talk] Online Sentiment and Topic Dynamics Tracking from Social Media

[ September 11, 2012; 4:00 pm to 5:00 pm. ] Location: MPEB 1.20 Title: Online Sentiment and Topic Dynamics Tracking from Social Media Speaker: Dr. Yulan He, Knowledge Media Institute, The Open University, UK Abstract: Social media data are produced continuously by a large and uncontrolled number of users. The dynamic nature of such data requires the sentiment and topic analysis model to be also dynamically updated, capturing the most recent language [...]