Search

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.ipm.2009.11.004 Byline: Jeongwoo Ko (a), Luo Si (b), Eric Nyberg (c) Keywords: Answer ranking; Answer merging; Question answering Abstract: Question answering (QA) aims at finding exact answers to a user's question from a large collection of documents. Most QA systems combine information retrieval with extraction techniques to identify a set of likely candidates and then utilize some ranking strategy to generate the final answers. This ranking process can be challenging, as it entails identifying the relevant answers amongst many irrelevant ones. This is more challenging in multi-strategy QA, in which multiple answering agents are used to extract answer candidates. As answer candidates come from different agents with different score distributions, how to merge answer candidates plays an important role in answer ranking. In this paper, we propose a unified probabilistic framework which combines multiple evidence to address challenges in answer ranking and answer merging. The hypotheses of the paper are that: (1) the framework effectively combines multiple evidence for identifying answer relevance and their correlation in answer ranking, (2) the framework supports answer merging on answer candidates returned by multiple extraction techniques, (3) the framework can support list questions as well as factoid questions, (4) the framework can be easily applied to a different QA system, and (5) the framework significantly improves performance of a QA system. An extensive set of experiments was done to support our hypotheses and demonstrate the effectiveness of the framework. All of the work substantially extends the preliminary research in . A probabilistic framework for answer selection in question answering. In: Proceedings of NAACL/HLT. Author Affiliation: (a) Search Quality Group, Google Inc., Mountain View, CA 94043, USA (b) Department of Computer Science, Purdue University, Lafayette, IN 47907, USA (c) School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213, USA Article History: Received 21 February 2008; Revised 9 November 2009; Accepted 11 November 2009

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.ipm.2005.09.005 Byline: Nicholas Lester (a), Justin Zobel (a), Hugh Williams (b) Keywords: Text indexing; Search engines; Index construction; Index update Abstract: Search engines and other text retrieval systems use high-performance inverted indexes to provide efficient text query evaluation. Algorithms for fast query evaluation and index construction are well-known, but relatively little has been published concerning update. In this paper, we experimentally evaluate the two main alternative strategies for index maintenance in the presence of insertions, with the constraint that inverted lists remain contiguous on disk for fast query evaluation. The in-place and re-merge strategies are benchmarked against the baseline of a complete re-build. Our experiments with large volumes of web data show that re-merge is the fastest approach if large buffers are available, but that even a simple implementation of in-place update is suitable when the rate of insertion is low or memory buffer size is limited. We also show that with careful design of aspects of implementation such as free-space management, in-place update can be improved by around an order of magnitude over a naA[macron]ve implementation. Author Affiliation: (a) School of Computer Science and Information Technology, RMIT University, Victoria 3001, Australia (b) MSN Search, Microsoft Corporation, One Microsoft Way, Redmond, WA 98052, United States Article History: Received 14 April 2005; Accepted 26 September 2005 Article Note: (footnote) [star] This paper incorporates and extends material from 'In-place versus re-build versus re-merge: Index maintenance strategies for text retrieval systems', N. Lester, J. Zobel, and H.E. Williams, Australasian Computer Science Conference, Dunedin, New Zealand, 2004.

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.ipm.2005.03.006 Byline: Jane Reid (a), Mounia Lalmas (a), Karen Finesilver (b), Morten Hertzum (c) Keywords: Structured document retrieval; Focussed retrieval; Best entry points; User studies Abstract: Structured document retrieval makes use of document components as the basis of the retrieval process, rather than complete documents. The inherent relationships between these components make it vital to support users' natural browsing behaviour in order to offer effective and efficient access to structured documents. This paper examines the concept of best entry points, which are document components from which the user can browse to obtain optimal access to relevant document components. In particular this paper investigates the basic characteristics of best entry points. Author Affiliation: (a) Computer Science, Queen Mary University of London, London E1 4NS, UK (b) Open and Distance Learning Unit, Queen Mary University of London, London E1 4NS, UK (c) Computer Science, Roskilde University, DK-4000, Denmark Article History: Received 27 September 2004; Accepted 3 March 2005

Byline: Dustin Walsh, Chad Livengood and Kirk Pinho On paper, Amazon Inc.'s quest for a second headquarters in North America employing 50,000 people in the next 15 years looks like [...]

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.ipm.2009.02.002 Byline: Alexandros Nanopoulos (a), Dimitrios Rafailidis (b), Maria M. Ruxanda (c), Yannis Manolopoulos (b) Keywords: Multimedia information retrieval; Audio input/output; Sound and music computing; Web search Abstract: Nowadays we have a proliferation of music data available over the Web. One of the imperative challenges is how to search these vast, global-scale musical resources to find preferred music. Recent research has envisaged the notion of music search engines (MSEs) that allow for searching preferred music over the Web. In this paper, we examine the growing research topic of MSEs, and provide potential specifications to follow and challenges to face. Author Affiliation: (a) Hildesheim University, Institute of Computer Science, Germany (b) Department of Informatics, Aristotle University of Thessaloniki, Greece (c) Department of Computer Science, Aalborg University, Denmark Article History: Received 8 April 2008; Revised 2 February 2009; Accepted 3 February 2009

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.cose.2008.11.008 Byline: Jen-Ho Yang (a), Chin-Chen Chang (a)(b) Abstract: Recently, remote user authentication schemes are implemented on elliptic curve cryptosystem (ECC) to reduce the computation loads for mobile devices. However, most remote user authentication schemes on ECC are based on public-key cryptosystem, in which the public key in the system requires the associated certificate to prove its validity. Thus, the user needs to perform additional computations to verify the certificate in these schemes. In addition, we find these schemes do not provide mutual authentication or a session key agreement between the user and the remote server. Therefore, we propose an ID-based remote mutual authentication with key agreement scheme on ECC in this paper. Based upon the ID-based concept, the proposed scheme does not require public keys for users such that the additional computations for certificates can be reduced. Moreover, the proposed scheme not only provides mutual authentication but also supports a session key agreement between the user and the server. Compared with the related works, the proposed scheme is more efficient and practical for mobile devices. Author Affiliation: (a) Department of Computer Science and Information Engineering, National Chung Cheng University, 160 San-Hsing, Ming-Hsiung, Chiayi 621, Taiwan, ROC (b) Department of Information Engineering and Computer Science, Feng Chia University, 100 Wenhwa Rd., Seatwen, Taichung 40724, Taiwan, ROC Article History: Received 21 August 2008; Accepted 26 November 2008

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.ipm.2004.11.003 Byline: ChengXiang Zhai (a), John Lafferty (b) Keywords: Retrieval models; Risk minimization; Statistical language models; Bayesian decision theory Abstract: This paper presents a probabilistic information retrieval framework in which the retrieval problem is formally treated as a statistical decision problem. In this framework, queries and documents are modeled using statistical language models, user preferences are modeled through loss functions, and retrieval is cast as a risk minimization problem. We discuss how this framework can unify existing retrieval models and accommodate systematic development of new retrieval models. As an example of using the framework to model non-traditional retrieval problems, we derive retrieval models for subtopic retrieval, which is concerned with retrieving documents to cover many different subtopics of a general query topic. These new models differ from traditional retrieval models in that they relax the traditional assumption of independent relevance of documents. Author Affiliation: (a) Department of Computer Science, University of Illinois at Urbana-Champaign, Illinois, United States (b) School of Computer Science, Carnegie Mellon University, Pittsburgh, PA, United States Article History: Accepted 12 November 2004

This white paper - 'The Benefits of Using Rigorously Tested Routines from Numerical Libraries--Computer Science Edition' speaks to matters affecting computationally-intensive application development. To the extent that software engineers and [...]

Tactile instruction is crucial to learning certain tasks, but unlike video and audio, touch is difficult to record and transfer. Now researchers from MIT's Computer Science and Artificial Intelligence Laboratory [...]

Byline: admin Main photo courtesy DepositPhotos. Artificial Intelligence has made its way into college catalogs across the country, and New Orleans-area universities are making their own investments in AI or [...]

'The rise of AI presents an opportunity for executives in every industry to differentiate and defend their businesses. But implementing a company-wide AI strategy is challenging, especially for legacy enterprises.' [...]

Yolanda Gil, AAAI President, Rao Kambhampati, AAAI Past President and Awards Committee Chair, and Bart Selman, AAAI President-Elect, presented the AAAI Awards in January at AAAI-19 in Honolulu. 2019 Feigenbaum [...]

We've come a long way since the first 3-D-printed item came to us by way of an eye wash cup, to now being able to rapidly fabricate things like car [...]

* Former president of the Association for the Advancement of Artificial Intelligence, and professor at the Massachusetts Institute of Technology Patrick Henry Winston, died July 19, 2019. In this article, [...]

New technologies based on artificial intelligence (AI) will transform not only the legal information world but also the management of law libraries and law firms. Al has moved from theoretical [...]

THE SILICON VALLEY GENDER NARRATIVE IS FIRMLY ENTRENCHED: WOMEN PONT GET COMPUTER SCIENCE DEGREES AND DON'T START SUCCESSFUL TECH COMPANIES. DON'T TELL THERESE TUCKER ANY OF THAT ONE EVENING IN [...]

Who: James Kuffner Titles: Director, chief digital officer, Toyota Motor Corp.; CEO, Toyota Research Institute u Advanced Development Age: 49 Joined Toyota: 2016 Google career: 2009-16, engineer on original self-driving [...]

For better or worse, the IoT will bring the business models that run the internet into the rest of the world IN JULY THE Bank of England announced that its [...]

Byline: Michelle Russell A team of researchers at Massachusetts Institute of Technology (MIT) is developing a special dye system that could allow consumers to change the colour and design of [...]

Byline: Beth Wright Researchers in the US have developed a new system and design tool for programming high-tech knitting machines and customising knitted patterns and shapes - in a move [...]