Download Advances in Information Retrieval: 36th European Conference by Maarten de Rijke, Tom Kenter, Arjen P. de Vries, ChengXiang PDF

By Maarten de Rijke, Tom Kenter, Arjen P. de Vries, ChengXiang Zhai, Franciska de Jong, Kira Radinsky, Katja Hofmann

This e-book constitutes the lawsuits of the thirty sixth ecu convention on IR learn, ECIR 2014, held in Amsterdam, The Netherlands, in April 2014.

The 33 complete papers, 50 poster papers and 15 demonstrations awarded during this quantity have been conscientiously reviewed and chosen from 288 submissions. The papers are equipped within the following topical sections: overview, advice, optimization, semantics, aggregation, queries, mining social media, electronic libraries, potency, and data retrieval concept. additionally incorporated are three instructional and four workshop presentations.

Show description

Read or Download Advances in Information Retrieval: 36th European Conference on IR Research, ECIR 2014, Amsterdam, The Netherlands, April 13-16, 2014. Proceedings PDF

Similar human-computer interaction books

Robotic Fabrication in Architecture, Art and Design 2014

Robot automation has turn into ubiquitous within the sleek production panorama, spanning an overpowering diversity of methods and applications-- from small scale force-controlled grinding operations for orthopedic joints to giant scale composite production of plane fuselages. clever factories, seamlessly associated through business networks and sensing, have revolutionized mass construction, taking into consideration clever, adaptive production tactics throughout a wide spectrum of industries.

Play: A Theory of Learning and Change

This e-book examines the query of why ‘play’ is a contented and benevolent verb in youth, but a subjective label of behaviour in maturity. It stories the transformation of the definitely labelled time period ‘child’s play’, used to consult our early years, into an aberrance or deviation from general social relationships in later existence, after we converse of taking part in up or being silly.

Extra info for Advances in Information Retrieval: 36th European Conference on IR Research, ECIR 2014, Amsterdam, The Netherlands, April 13-16, 2014. Proceedings

Sample text

ECIR 2010. LNCS, vol. 5993, pp. 153–165. : On aggregating labels from multiple crowd workers to infer relevance of documents. , Silvestri, F. ) ECIR 2012. LNCS, vol. 7224, pp. 182–194. : Rank Correlation Methods, 1st edn. : Automatic ranking of information retrieval systems using data fusion. : Ranking retrieval systems without relevance judgments. In: Proc. 24th Annual International ACM SIGIR, pp. : Report on the need for and provision of an ‘ideal’ test collection. Tech. : Estimating average precision with incomplete and imperfect judgments.

From the plots and tables, we can also see that the DFI models were not the fairest models, but they are all reasonably effective (and the best performing model on T123, and second best on AQ and WT) with the DFIb term weighting scheme performing the best out of all DFI models. IDF based models which don’t explicitly perform document length normalisation and the other models, which do. Within this second group, however, the relationship is bit more complicated, as can be seen from the WT plot in Figure 1.

This is the Hedge loss function defined by Aslam et al. [1]. In our experiments, we select documents to be judged only from the TREC judgment pool, as we do not know the correct judgment for documents outside the pool. Both P1 and P2 reflect the fact that the distinction between positive judgments and unknown judgments is consequential for Mean Average Precision, our evaluation measure, while the distinction between negative judgments and unknown judgments is inconsequential for that measure. Our use of standard deviation in P2 reflects our intuition that a greater diversity of system responses for the same document might be an indication of poorly estimated pseudo-judgments, and also of documents likely to be discriminative between systems.

Download PDF sample

Rated 4.59 of 5 – based on 22 votes