University of Avignon

About: University of Avignon is a education organization based out in Avignon, Provence-Alpes-Côte-d'Azur, France. It is known for research contribution in the topics: Population & Speaker recognition. The organization has 1526 authors who have published 3766 publications receiving 88928 citations.

Estimating ungauged catchment flows from Lake Tana floodplains, Ethiopia: an isotope hydrological approach.

Abstract: The isotope balance approach, which used 18O content of waters, has been used as an independent tool to estimate inflow to Lake Tana of surface water flows from ungauged catchment of Lake Tana (50% of the total area) and evaporative water loss in the vast plains adjoining the lake. Sensitivity analysis has been conducted to investigate the effects of changes in the input parameters on the estimated flux. Surface water inflow from ungauged catchment is determined to be in the order of 1.698×109 m3a−1. Unaccounted water loss from the lake has been estimated at 454×106 m3a−1 (equivalent to 5% of the total via surface water). Since the lake is water tight to groundwater outflow, the major error introduced into the water balance computation is related to evaporative water loss in water from the flood plains. If drained, the water which is lost to evaporation can be used as an additional water resource for socio-economic development in the region (tourism, agriculture, hydropower, and navigation). Hydrological ...

Diversity of spore germination in response to inosine and L-alanine and its interaction with NaCl and pH in the Bacillus cereus group.

Abstract: Aims: Our aim was to assess the diversity of the nutrient germination response of Bacillus cereus spores. Methods and Results: B. cereus spore germination was monitored by decrease in optical density using a Bioscreen C analyser in response to the major germinant substances inosine and l-alanine. Spores of a set of 12 strains taken to illustrate the diversity of the B. cereus group showed ranging germination capacities. Two strains never germinated in the presence of l-alanine, at any of the germinant concentrations tested. Both the extent and rate of spore germination were affected by low pH and high NaCl concentration, but differently according to the strain. Conclusions: A broad diversity was observed in nutrient-triggered spore germination among the members of the B. cereus group. Spore germination of some strains occurred at low concentrations of inosine or l-alanine, suggesting high receptor sensitivity to germinants. The activity of these receptors was also affected by pH or high NaCl concentration. Significance and Impact of the Study: The greater ability of some strains to germinate in response to l-alanine and inosine is one criterion among others for B. cereus strain selection in food processing or storage studies, before confirmation in complex food or laboratory media. The diversity in response to germinants found among the B. cereus strains suggests a differential expression and (or) absence of some germination genes involved in the response, mainly to l-alanine.

Topology Optimization of Electric Motor Using Topological Derivative for Nonlinear Magnetostatics

Abstract: We aim at finding an optimal design for an interior permanent magnet electric motor by means of a sensitivity-based topology optimization method. The gradient-based ON/OFF method has been successfully applied to optimization problems of this form. We show that this method can be improved by considering the mathematical concept of topological derivatives (TDs). TDs for optimization problems constrained by linear partial differential equations (PDEs) are well understood, whereas little is known about TDs in combination with nonlinear PDE constraints. We derive the TD for an optimization problem constrained by the equation of nonlinear 2-D magnetostatics, illustrate its advantages over the sensitivities used in the ON/OFF method, and show numerical results for the optimization of an interior permanent magnet electric motor obtained by a level-set algorithm, which is based on the TD.

Severe drought-induced community tolerance to heat wave. An experimental study on soil microbial processes

Abstract: The purposes of this study were to identify the influence of a severe drought period on the impact of a subsequent heat–drought disturbance on the microbial community of a Mediterranean agricultural soil and particularly to highlight the long-term effects on the microbial catabolic profiles. We performed an experiment in microcosms and applied the MicroResp™ method on soil microbial communities. A 21-day combined heat–drought disturbance had less impact on soil microbial communities pre-exposed to a 73-day severe drought than on those that were not pre-exposed. These differences were observed not only for biomass and physiological traits (basal respiration, qCO2), but also for catabolic microbial structure evolution during the recovery time. These observations suggest that the physiological stress imposed by the initial severe drought changed the microbial catabolic structure or physiological state and favoured a portion of the microbial community best adapted to cope with the final heat–drought disturbance. Consequently, the initial severe drought may have induced a community tolerance to the subsequent heat wave. In this study, we also note that resilience was, more than resistance, an indicator of pre-exposure to stress. In the context of assessing the effects of extreme climatic events on soil microbial processes, these results suggest that future studies should take into account the historic stress of habitats and resilience parameters.

Improving dialogue classification using a topic space representation and a Gaussian classifier based on the decision rule

Abstract: In this paper, we study the impact of dialogue representations and classification methods in the task of theme identification of telephone conversation services having highly imperfect automatic transcriptions. Two dialogue representations are firstly compared: the classical Term Frequency-Inverse Document Frequency with Gini purity criteria (TF-IDF-Gini) method and the Latent Dirichlet Allocation (LDA) approach. We then propose to study an original classification method that takes advantage of the LDA topic space representation, highlighted as the best dialogue representation. To do so, two assumptions about topic representation led us to choose a Gaussian process (GP) based method. This approach is compared with a Support Vector Machine (SVM) classification method. Results show that the GP approach is a better solution to deal with the multiple theme complexity of a dialogue, no matter the conditions studied (manual or automatic transcriptions). We finally discuss the impact of the topic space reduction on the classification accuracy.