Le dialogue homme- machine en langue naturelle : un défi ?

The implementation of a natural language manmachine dialogue in a data base or knowledge base query system is a difficult task. We present hereafter three different approaches to this problem : dialog modeling, exemplified through a linguistic scheme and several computational implementations ; the cooperation principles in a dialog and their associated algorithms ; and, lastly, the artificial intelligence approach based on speech acts theory and the techniques for generating and recognizing plans. We conclude by assessing the possible use of current speech processing techniques and by listing still unsolved problems.     

Unusual attendance at Hospital Emergency Services for asthma and chronic obstructive pulmonary disease and SO2 air pollution in Cartagena (Spain)

BACKGROUND: The objective of this investigation is to determine the influence of air pollution (SO2 and particulate matter) on the existence of days of unusual attendance at emergency services for asthma and chronic obstructive pulmonary lung disease (COPD). A registry of attendances for these diseases and the daily mean concentrations of SO2 and particles were used (1989-1991). RESULTS: Unusual attendance days are establish by the fortnightly movable mean of the period that has that day as central one, under the hypothesis of a Poisson distribution with equal mean as the calculated one. In the logistic regression model, Odds Ratio (OR) between an excess of attendance of cases for asthma and SO2 levels greater than 80.60 m/m3 was 3.6 (CI95%: 1.1-11.7). For COPD, SO levels ten days before were introduced, and OR for SO2 levels more than 56.5m/m was 4.7 (CI95%: 1.5-15.1). CONCLUSIONS: High SO2 levels are related with the appearances of days with an excessive use of emergencies for Asma and EPOC.     

VITALITY OF CIDER YEAST GROWN MICRO-AEROBICALLY WITH ADDED ETHANOL,BUTAN-J-OL OR ISO-BUTANOL

Vitality and viability of an alcohol-tolerant wine yeast, used in cider production, were assessed after exposure to alkanol(s) during growth. Criteria employed were: methylene blue reduction, ability to form colonies on yeast extract-peptone-glucose agar medium, glucose driven proton efflux rates (“acidification power”), fermentative (CO₂ output) rates and adenylate energy charge values. We also monitored the maintenance of transmembrane electrochemical potential across the plasma membrane as measured by flow cytometry and by scanning confocal laser microscopy of oxonol dye exclusion. Growth rates were diminished by a third by 7.5% (v/v) added ethanol, 1% butan-1-ol or 1.4% iso-butanol. Exposure to 10% (v/v) ethanol gave 16% loss of “viability”, as measured by methylene blue reduction, during the first 20 h of growth. For 1% butan-1-ol, 50% loss of “viability” occurred over 40 h, whereas a similar effect of iso-butanol took 55 h. Adenylate charge values were high (<0.8) in growing cultures, remained high in early stationary phase but declined to 0.4 after 115 h. These values were hardly affected by 5 or 7.5% (v/v) ethanol whereas 10% or 15% (v/v) ethanol gave values of 0.58 and 0.16 after only 5 h exposure. 1% butan-1-ol or iso-butanol decreased adenylate charge values to a greater extent than 10% (v/v) ethanol, with the straight chain alcohol the more potent. Oxonol exclusion indicated that the vast majority of colls with greatly diminished vitality have maintained the plasma membrane potential values required to retain viability, despite extensive exposure to alkanol(s). Thus loss of ability to reduce methylene blue indicates diminished vitality but is not a reliable index of loss of viability. “Acidification power” was a more sensitive indicator of vitality than adenylate charge values. When mixtures of C₂ + C₄ alcohols were employed effects were generally additive rather than synergistic.     

Parallel multipole implementation of the generalized Helmholtz decomposition for solving viscous flow problems

The evaluation of a domain integral is the dominant bottleneck in the numerical solution of viscous flow problems by vorticity methods, which otherwise demonstrate distinct advantages over primitive variable methods. By applying a Barnes–Hut multipole acceleration technique, the operation count for the integration is reduced from O(N²) to O(NlogN), while the memory requirements are reduced from O(N²) to O(N). The algorithmic parameters that are necessary to achieve such scaling are described. The parallelization of the algorithm is crucial if the method is to be applied to realistic problems. A parallelization procedure which achieves almost perfect scaling is shown. Finally, numerical experiments on a driven cavity benchmark problem are performed. The actual increase in performance and reduction in storage requirements match theoretical predictions well, and the scalability of the procedure is very good. Copyright © 2003 John Wiley Sons, Ltd.     

Design of adaptive feedforward algorithms using internal model equivalence

The paper investigates the design of adaptive feedforward cancellation (AFC) algorithms with sinusoidal regressors for repetitive control. Such adaptive algorithms are equivalent to linear controllers based on the internal model principle (IMP). Using this equivalence and root locus rules, the phase advance of the regressor of the adaptive algorithm can be chosen to maximize the phase margin at low gains. A surprising result is that selecting the optimal phase advance is equivalent to placing a zero in the open right half-plane in certain cases. A complete design and analysis for the compensation of a single-frequency periodic disturbance is done. A new variation of the AFC algorithm is also developed in which the adaptive portion acts in parallel with a feedthrough term. the IMP equivalent of this algorithm has two zeros instead of one. Analysis and simulation show this method to have superior convergence and robustness properties when compared with the method having no feedthrough term. Discrete time versions of the algorithms are briefly considered.     

The active site of Trichoderma reesei cellobiohydrolase II: the role of tyrosine 169

Trichoderma reesei cellobiohydrolase II (CBHII) is an exoglucanasecleaving primarily cellobiose units from the non-reducing endof cellulose chains. The ß-l,4 glycosidic bond iscleaved by acid catalysis with an aspartic acid, D221, as thelikely proton donor, and another aspartate, D175, probably ensuringits protonation and stabilizing charged reaction intermediates.The catalytic base has not yet been identified experimentally.The refined crystal structure of CBHII also shows a tyrosineresidue, Y169, located close enough to the scissile bond tobe involved in catalysis. The role of this residue has beenstudied by introducing a mutation Y169F, and analysing the kineticand binding behaviour of the mutated CBHII. The crystal structureof the mutated enzyme was determined to 2.0 Å resolutionshowing no changes when compared with the structure of nativeCBHII. However, the association constants of the mutant enzymefor cellobiose and cellotriose are increased threefold and for4-methylumbelliferyl cellobioside over 50-fold. The catalyticconstants towards cellotriose and cellotetraose are four timeslower for the mutant. These data suggest that Y169, on interactingwith a glucose ring entering the second subsite in a narrowtunnel, helps to distort the glucose ring into a more reactiveconformation. In addition, a change in the pH activity profilewas observed. This indicates that Y169 may have asecond rolein the catalysis, namely to affect the protonation state ofthe active site carboxylates, D175 and D221.     

Custody evaluation practices: A survey of experienced professionals (revisited).

W. G. Keilin and L. J. Bloom (1986) explored 70 issues related to child custody evaluation practices. The current study replicates Keilin and Bloom's study and looks at an additional 42 items. Two hundred and one psychologists from 39 states were surveyed about 112 aspects of child custody evaluation practices. The analysis was divided into evaluation practices, sole-joint custody decision making, and recommendations. Comparisons between the findings of this study and those of Keilin and Bloom are made. The current practice of child custody evaluations is reported. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Intraarterial urokinase for acute native arterial occlusion of the limbs

Since 1988, 49 limbs of 47 patients underwent intraarterial urokinase infusion for acute native artery occlusion. The time from the onset of ischemic symptoms ranged from 1 to 45 days (mean = 17.5). The arterial sectors involved were femoropopliteal in 32 cases, followed by aortoiliac in 13 cases, distal in three cases, and subclavian in one case. Treatment consisted of placing a catheter in the clot and the infusion of 4400 U/kg in 20', followed by a series of 4400 U/kg weight/hour during 6 hours. Clinical evaluation, hemodynamic and coagulation parameters, and angiographical changes were assessed periodically. Infusion time ranged from 6 to 24 hours (mean = 13.2 hours). Improvement of ischemic was achieved in 43 (87.75%) patients. In five patients (12.25%) there was no improvement. Total immediate lysis was achieved in 35 cases (71.5%), and among them, 13 patients (26%) required no associated treatment, 16 (48%) underwent PTA, and four (12%) had surgery of underlying peripheral aneurysms revealed after thrombolysis. Partial lysis was achieved in 13 cases (26.5%), that was enough in four of them, but the remaining nine required further treatment (four PTA, and five arterial surgery). In one case no lysis was achieved, and arterial surgery was carried out. No mortality was recorded, and major complications included one upper gastrointestinal bleeding, and one cerebral hematoma. Late follow-up of successfully treated patients who did not require further surgery shows a cumulative patency rate of 81% at 24 months.     

Hierarchical approach to modelling of liquid-crystalline polymers

Summary Liquid-crystalline polymers exhibit fascinating structure at many size scales. This paper describes how computer models at different size scales can be linked together by a hierarchical approach to give a better understanding of the properties and behaviour of these materials. The atomic scale is considered first, where semi-empirical molecular orbital techniques are used to calculate the torsional energy functions associated with rotating the backbone bonds of the polymer. Secondly, entire chains are simulated using a Monte Carlo technique based on the torsional energy functions to deduce the persistence length of the polymer of interest. A theoretical relationship enables the Frank elastic constants to be determined from the persistence length. The elastic constants may then be used as input parameters for models both to predict microstructure and to help understand the role of defects in the shear flow of these materials. The hierarchical approach provides a meaningful framework within which data obtained from small-scale models are used to parameterise models at a larger scale.