[IR Series Talk @ 11:30 Friday] : Utility-based Information Distillation Over Temporally Sequenced Documents

[Grace Hui Yang]

Dear All,
    This is a gentle reminder that the IR Series talk will be starting 
at 11:30am  this Friday, NOT 12:00, which is the usual time. Please make 
a note for this.
    Thanks! 
Grace, Jaime, Jon

-------------------------------------------------------------------------------------------------------------------------------------------------

Grace Hui Yang wrote:
> Dear All,
> We are pleased to announce the upcoming IR series talk on Friday. The 
> talk will be given by Professor Yiming Yang. The talk is about 
> "Utility-based Information Distillation Over Temporally Sequenced 
> Documents".
>
> More Details:
>
> Time: 11:30am July 13 2007
> Location: NSH 1507
> (Food will be provided. Thanks for our sponsor Yahoo!)
>
> Speaker: Professor Yiming Yang
> Topic: Utility-based Information Distillation Over Temporally 
> Sequenced Documents".
>
> Abstract:
> This talk examines a new approach to information distillation over 
> temporally ordered documents, and proposes a
> novel evaluation scheme for such a framework. It combines the 
> strengths of and extends beyond conventional adaptive
> filtering, novelty detection and non-redundant passage ranking with 
> respect to long-lasting information needs
> (‘tasks’ with multiple queries). Our approach supports fine-grained 
> user feedback via highlighting of arbitrary
> spans of text, and leverages such information for utility optimization 
> in adaptive settings. For our experiments,
> we defined hypothetical tasks based on news events in the TDT4 corpus, 
> with multiple queries per task. Answer
> keys (nuggets) were generated for each query and a semiautomatic 
> procedure was used for acquiring rules that allow
> automatically matching nuggets against system responses. We also 
> propose an extension of the NDCG metric for
> assessing the utility of ranked passages as a combination of relevance 
> and novelty. Our results show encouraging utility
> enhancements using the new approach, compared to the baseline systems 
> without incremental learning or the novelty
> detection components.
>
> The paper is published in SIGIR 2007.
>
> See you there!
> Grace, Jaime, Jon
>