From huiyang at cs.cmu.edu Wed Jul 11 11:16:33 2007 From: huiyang at cs.cmu.edu (Grace Hui Yang) Date: Wed, 11 Jul 2007 11:16:33 -0400 Subject: [IR Series Talk @ 11:30 Friday] : Utility-based Information Distillation Over Temporally Sequenced Documents In-Reply-To: <46929DBF.9050304@cs.cmu.edu> References: <46929DBF.9050304@cs.cmu.edu> Message-ID: <4694F451.7080609@cs.cmu.edu> 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 > From huiyang at cs.cmu.edu Tue Jul 17 16:06:53 2007 From: huiyang at cs.cmu.edu (Grace Hui Yang) Date: Tue, 17 Jul 2007 16:06:53 -0400 Subject: [IR Series] : Structured Retrieval for Question Answering @ 20/07/07 Message-ID: <469D215D.3070906@cs.cmu.edu> Dear All, We are pleased to announce the IR series talk on this Friday. The talk will be given by Matthew W. Bilotti. The talk is about "Structured Retrieval for Question Answering". Lunch will be provided by our sponsor Yahoo!. See you there! Time: 12:00 noon Date: July 20 2007 (Friday) Location: NSH 1109 Speaker: Matthew W. Bilotti Topic: Structured Retrieval for Question Answering Abstract: Bag-of-words retrieval is popular among Question Answering (QA) system developers, but it does not support constraint checking and ranking on the linguistic and semantic information of interest to the QA system. We present an approach to retrieval for QA, applying structured retrieval techniques to the types of text annotations that QA systems use. We demonstrate that the structured approach can retrieve more relevant results, more highly ranked, compared with bag-of-words, on a sentence retrieval task. We also characterize the extent to which structured retrieval effectiveness depends on the quality of the annotations. The paper is published in SIGIR 2007. Here is a copy of the paper : http://www.cs.cmu.edu/~mbilotti/pubs/Bilotti:SIGIR07.pdf ---------- Grace, Jaime, Jon From jelsas+ at cs.cmu.edu Thu Jul 19 13:22:59 2007 From: jelsas+ at cs.cmu.edu (Jonathan Elsas) Date: Thu, 19 Jul 2007 13:22:59 -0400 Subject: Reminder : Structured Retrieval for Question Answering @ 20/07/07 In-Reply-To: <469D215D.3070906@cs.cmu.edu> References: <469D215D.3070906@cs.cmu.edu> Message-ID: <8F079240-8736-4473-97FE-2CCF0121DEE0@cs.cmu.edu> Reminder: the IR series will have a talk tomorrow. Matthew W. Bilotti will present "Structured Retrieval for Question Answering" Note: This talk is not in our usual room. NSH 1109 is in the Field Robotics Center. On Jul 17, 2007, at 4:06 PM, Grace Hui Yang wrote: > Dear All, > We are pleased to announce the IR series talk on this Friday. The > talk will be given by Matthew W. Bilotti. The talk is about > "Structured Retrieval for Question Answering". Lunch will be > provided by our sponsor Yahoo!. > > See you there! > > > Time: 12:00 noon > Date: July 20 2007 (Friday) > Location: NSH 1109 > > Speaker: Matthew W. Bilotti > Topic: Structured Retrieval for Question Answering > > Abstract: > Bag-of-words retrieval is popular among Question Answering > (QA) system developers, but it does not support constraint > checking and ranking on the linguistic and semantic > information of interest to the QA system. We present an > approach to retrieval for QA, applying structured retrieval > techniques to the types of text annotations that QA systems > use. We demonstrate that the structured approach can retrieve > more relevant results, more highly ranked, compared > with bag-of-words, on a sentence retrieval task. We also > characterize the extent to which structured retrieval effectiveness > depends on the quality of the annotations. > > > The paper is published in SIGIR 2007. Here is a copy of the paper : > http://www.cs.cmu.edu/~mbilotti/pubs/Bilotti:SIGIR07.pdf > > ---------- > Grace, Jaime, Jon > From jaime at cs.cmu.edu Fri Jul 20 10:44:59 2007 From: jaime at cs.cmu.edu (Jaime Arguello) Date: Fri, 20 Jul 2007 10:44:59 -0400 Subject: (Today!) Structured Retrieval for Question Answering @ 20/07/07 In-Reply-To: <469D215D.3070906@cs.cmu.edu> References: <469D215D.3070906@cs.cmu.edu> Message-ID: <46A0CA6B.8090408@cs.cmu.edu> All, Please take notice of the unusual meeting place for today's talk, NSH 1109. This room is located inside the Field Robotics Center, on the 1st floor of Newell-Simon Hall. Thank you, Jaime Arguello > Dear All, > We are pleased to announce the IR series talk on this Friday. The talk > will be given by Matthew W. Bilotti. The talk is about "Structured > Retrieval for Question Answering". Lunch will be provided by our > sponsor Yahoo!. > > See you there! > > > Time: 12:00 noon > Date: July 20 2007 (Friday) > Location: NSH 1109 > > Speaker: Matthew W. Bilotti > Topic: Structured Retrieval for Question Answering > > Abstract: > Bag-of-words retrieval is popular among Question Answering > (QA) system developers, but it does not support constraint > checking and ranking on the linguistic and semantic > information of interest to the QA system. We present an > approach to retrieval for QA, applying structured retrieval > techniques to the types of text annotations that QA systems > use. We demonstrate that the structured approach can retrieve > more relevant results, more highly ranked, compared > with bag-of-words, on a sentence retrieval task. We also > characterize the extent to which structured retrieval effectiveness > depends on the quality of the annotations. > > > The paper is published in SIGIR 2007. Here is a copy of the paper : > http://www.cs.cmu.edu/~mbilotti/pubs/Bilotti:SIGIR07.pdf > > ---------- > Grace, Jaime, Jon >