Edinburgh Napier University

About: Edinburgh Napier University is a education organization based out in Edinburgh, United Kingdom. It is known for research contribution in the topics: Population & Context (language use). The organization has 2665 authors who have published 6859 publications receiving 175272 citations.

The water vapour sorption properties of Sitka spruce determined using a dynamic vapour sorption apparatus

Abstract: The water vapour sorption properties of Sitka spruce (Picea abies) have been investigated over a range of temperatures (14.2–43.8°C) using a dynamic vapour sorption apparatus. The sorption kinetics behaviour was evaluated using the parallel exponential kinetics model which has been found to give very accurate fits to the data in studies of foodstuffs or plant fibres, but has not been previously applied to sorption studies with wood. Both the adsorption and desorption kinetics curve can be deconvoluted into a fast and slow exponential process. Under conditions of adsorption, the fast process appears to be associated with the formation of monolayer water (determined using the Hailwood Horrobin model) up to a relative humidity of 20%. Under desorption, there is no clear differentiation between fast and slow processes. The area bounded by the sorption hysteresis loop reduced as the temperature at which the isotherm was measured increased, due to movement of the desorption curve only, with the adsorption curve remaining the same at all temperatures. This behaviour is consistent with sorption processes taking place on nanoporous glassy solids below the glass transition temperature. The heat of wetting was determined from the temperature dependence of the desorption isotherms by using the Clausius–Clapeyron equation, yielding results that are consistent with literature values. However, doubts are raised in this paper as to the applicability of using the Clausius–Clapeyron equation for analyses of this type.

Stochastic Answer Networks for Machine Reading Comprehension

Abstract: We propose a simple yet robust stochastic answer network (SAN) that simulates multi-step reasoning in machine reading comprehension. Compared to previous work such as ReasoNet which used reinforcement learning to determine the number of steps, the unique feature is the use of a kind of stochastic prediction dropout on the answer module (final layer) of the neural network during the training. We show that this simple trick improves robustness and achieves results competitive to the state-of-the-art on the Stanford Question Answering Dataset (SQuAD), the Adversarial SQuAD, and the Microsoft MAchine Reading COmprehension Dataset (MS MARCO).

Electric discharge in water as a source of UV radiation, ozone and hydrogen peroxide

Abstract: Results are presented from investigations of multispark electric discharge in water excited along multielectrode metal-dielectric systems with gas supply into the interelectrode gaps. The intensity distribution of discharge radiation in the region covering the biologically active soft UV (190≤λ≤430 nm) has been determined and the absolute number of quanta in this wavelength interval has been measured. The potentiality of the slipping surface discharge in water for its disinfection is analysed. The energy expenditure for water cleansing is estimated to be as low as ~10-4 kWh l-1.