Using triangulation to identify word senses

Word sense disambiguation is the task of determining which sense of a word is intended from its context. Previous methods have found the lack of training data and the restrictiveness of dictionaries' choices of senses to be major stumbling blocks. A robust novel algorithm is presented that uses multiple dictionaries, the Internet, clustering and triangulation to attempt to discern the most useful senses of a given word and learn how they can be disambiguated. The algorithm is explained, and some promising sample results are given.

Developments in near infrared spectroscopy

Vibrational spectroscopy at high excitation is an important research frontier for two reasons. Firstly, the near infrared is proving to be an important area for the analytical applications of spectroscopy, and we would therefore like to understand how the spectra we observe relate to the molecular structure of the absorbing species. Secondly, there is a fundamental interest in understanding molecular dynamics and energy flow within a polyatomic molecule at high excitation, because this is the boundary between spectroscopy and chemistry through which we try to understand the details of a chemical reaction. In this presentation I shall survey recent progress in this field.

Identifying problematic classes in text classification

Real-world text classification tasks often suffer from poor class structure with many overlapping classes and blurred boundaries. Training data pooled from multiple sources tend to be inconsistent and contain erroneous labelling, leading to poor performance of standard text classifiers. The classification of health service products to specialized procurement classes is used to examine and quantify the extent of these problems. A novel method is presented to analyze the labelled data by selectively merging classes where there is not enough information for the classifier to distinguish them. Initial results show the method can identify the most problematic classes, which can be used either as a focus to improve the training data or to merge classes to increase confidence in the predicted results of the classifier.

Classification in e-procurement

Three coupled knowledge transfer partnerships used pattern recognition techniques to produce an e-procurement system which, the National Audit Office reports, could save the National Health Service £500 m per annum. An extension to the system, GreenInsight, allows the environmental impact of procurements to be assessed and savings made. Both systems require suitable products to be discovered and equivalent products recognised, for which classification is a key component. This paper describes the innovative work done for product classification, feature selection and reducing the impact of mislabelled data.

Adsorption techniques in drinking water treatment : papers and discussion from the NATO/CCMS symposium held in Reston, Virginia, USA, April 30 through May 2, 1979 /

Mode of access: Internet.

The administration of emergency projects. (For administrative use)

Mode of access: Internet.

The life and Epistles of St. Paul /

Contiene: Vol. I (XXVII, 551 p., [3] h. de lám., [4] h. de map. pleg.) -- Vol. II (VI, 692 p., [3] h. de lám., [2] h. de map. pleg.)

Paul of Tarsus : an inquiry into the times and the gospel of the Apostle of the Gentiles /

Mode of access: Internet.

Observation of the suppression of the flux of cosmic rays above 4x10(19) eV

The energy spectrum of cosmic rays above 2.5 x 10(18) eV, derived from 20 000 events recorded at the Pierre Auger Observatory, is described. The spectral index gamma of the particle flux, J proportional to E(-gamma), at energies between 4 x 10(18) eV and 4 x 10(19) eV is 2.69 +/- 0.02(stat) +/- 0.06(syst), steepening to 4.2 +/- 0.4(stat) +/- 0: 06 (syst) at higher energies. The hypothesis of a single power law is rejected with a significance greater than 6 standard deviations. The data are consistent with the prediction by Greisen and by Zatsepin and Kuz'min.

Upper limit on the diffuse flux of ultrahigh energy tau neutrinos from the Pierre Auger Observatory

The surface detector array of the Pierre Auger Observatory is sensitive to Earth-skimming tau neutrinos that interact in Earth's crust. Tau leptons from nu(tau) charged-current interactions can emerge and decay in the atmosphere to produce a nearly horizontal shower with a significant electromagnetic component. The data collected between 1 January 2004 and 31 August 2007 are used to place an upper limit on the diffuse flux of nu(tau) at EeV energies. Assuming an E(nu)(-2) differential energy spectrum the limit set at 90% C. L. is E(nu)(2)dN(nu tau)/dE(nu) < 1: 3 x 10(-7) GeV cm(-2) s(-1) sr(-1) in the energy range 2 x 10(17) eV< E(nu) < 2 x 10(19) eV.

The Genome Sequence of Taurine Cattle: A Window to Ruminant Biology and Evolution

To understand the biology and evolution of ruminants, the cattle genome was sequenced to about sevenfold coverage. The cattle genome contains a minimum of 22,000 genes, with a core set of 14,345 orthologs shared among seven mammalian species of which 1217 are absent or undetected in noneutherian (marsupial or monotreme) genomes. Cattle-specific evolutionary breakpoint regions in chromosomes have a higher density of segmental duplications, enrichment of repetitive elements, and species-specific variations in genes associated with lactation and immune responsiveness. Genes involved in metabolism are generally highly conserved, although five metabolic genes are deleted or extensively diverged from their human orthologs. The cattle genome sequence thus provides a resource for understanding mammalian evolution and accelerating livestock genetic improvement for milk and meat production.