Semantic orientation for polarity classification in Spanish reviews

Until now most of the published methods for polarity classification have been applied to English texts, but other languages are becoming increasingly important. This paper presents a new resource for the Spanish sentiment analysis research community. We have generated a new lexicon by translating into Spanish the Bin Liu English Lexicon. In order to assess the validity of the proposed lexicon a set of experiments on a Spanish review corpus are presented. In addition, the resource presented is compared with another existing Spanish lexicon. The results show that our resource outperforms the currently available Spanish lexicon for sentiment analysis.     

RP-HPLC peptide profiling of cheese extracts: A study of sources of variation, repeatability and reproducibility

Peptide profiling of cheese extracts by RP-HPLC is widely used to study cheese quality and authenticity. Repeatability, reproducibility and the contribution of differences in sample/treatment, extraction, and the effect of the chromatographic run was studied. Chromatograms (108) of pH 4.6-soluble extracts from 180 d Cheddar cheeses were used to evaluate the precision of the method and to quantify the effect of treatment and replicate cheese trial. Repeatability and reproducibility coefficient of variations of peaks/RT classes varied widely and no clear correlation was found between peak size and reproducibility. ANOVA of Principal Component Scores identified significant differences amongst cheeses and a trial effect. Partial Least Square Discriminant Analysis identified variables related to cheese treatment. Statistical process control methods were used to study the variability of replicate injections over the life of a single column showing the discriminatory power of RP-HPLC and giving possibilities to improve both process control and process optimisation.     

Influence of some oenological practices on the aromatic and sensorial characteristics of white Verdejo wines

Changes in the aromatic composition as well as sensory characteristics in Verdejo white wines were analysed based on two factors: the winemaking methodology and the storing time of wine in bottles. The volatile components were determined by GLC‐MS, and the sensory profile was designed and assessed according to the ISO 11035 standard. The results showed that when wines were made in oak barrels, either completely or partially, which means the wines were in contact with the lees, the levels of 1‐octanol, ethyl heptanoate and ethyl decanoate were significantly affected (P < 0.05); the softness sensation was also influenced (P < 0.05). However, the amount of time the wines were stored in bottles significantly affected (P < 0.05) the levels of 1‐hexanol, ethyl heptanoate, ethyl octanoate, ethyl decanoate, hexyl acetate, isoamyl acetate and isoamyl lactate and also an odour note (tropical fruit). The compounds with higher OAV values belong to the groups of esters and fatty acids. For these reasons, the composition and the quality of the aroma of Verdejo white wines appear to be significantly affected both by use of oak barrels in winemaking and the time the wines are stored in bottles.     

FEM modelling and experimental characterization of microbeams in presence of residual stress

Out-of-plane bending tests are here used to experimentally validate some numerical models of microbeams actuated by the electric field. Out-of-plane bending microcantilevers and clamped–clamped microbeams often suffer the presence of residual strain and stress, respectively, which affect their static and dynamic behaviour and pull-in voltage. In case of microcantilever an accurate modelling has to include the effect of an initial curvature due to microfabrication process, while in double clamped microbeams constraints may impose a pre-loading caused by a tensile stress. So-called geometrical nonlinearity sometimes occurs, when microcantilever exhibits large displacement, or because of the mechanical coupling between axial and flexural behaviours in double clamped microbeams. Modelling this kind of nonlinearity is an additional goal of this study. Experiments demonstrated a good agreement with results of FEM approaches proposed. In the case of microbridges numerical models are used to identify the residual stress. A reverse analysis is implemented, the axial pre-stress is calculated by means of the measured pull-in voltage.     

An improved finite point method for tridimensional potential flows

At the local level, successful meshless techniques such as the Finite Point Method must have two main characteristics: a suitable geometrical support and a robust numerical approximation built on the former. In this article we develop the second condition and present an alternative procedure to obtain shape functions and their derivatives from a given cloud of points regardless of its geometrical features. This procedure, based on a QR factorization and an iterative adjust of local approximation parameters, allows obtaining a satisfactory minimization problem solution, even in the most difficult cases where usual approaches fail. It is known that high-order meshless constructions need to include a large number of points in the local support zone and this fact turns the approximation more dependent on the size, shape and spatial distribution of the local cloud of points. The proposed procedure also facilitates the construction of high-order approximations on generic geometries reducing their dependence on the geometrical support where they are based. Apart from the alternative solution to the minimization problem, the behaviour of high-order Finite Point approximations and the overall performance of the proposed methodology are shown by means of several numerical tests. Sergio Idelsohn is ICREA Research Professor at CIMNE.     

PTR-ToF-MS, A Novel, Rapid, High Sensitivity and Non-Invasive Tool to Monitor Volatile Compound Release During Fruit Post-Harvest Storage: The Case Study of Apple Ripening

In the present study, the potential of PTR-ToF-MS for addressing fundamental and technical post-harvest issues was tested on the non-destructive and rapid monitoring of volatile compound evolution in three apple cultivars (‘Golden Delicious’, ‘Braeburn’ and ‘Gold Rush’) during 25 days of post-harvest shelf life ripening. There were more than 800 peaks in the PTR-ToF-MS spectra of apple headspace and many of them were associated with relevant compounds. Besides the ion produced upon proton transfer, we used the ion at mass 28.031 (C₂H ₄ ⁺ ) produced by charge transfer from residual O ₂ ⁺ as a monitor for ethylene concentration. ‘Golden Delicious’ apples were characterised by higher ethylene emission rates than ‘Gold Rush’ and ‘Braeburn’, and quantitative comparison has been supported by two segment piecewise linear model fitting. Ester evolution during post-harvest ripening is strongly dependent on endogenous ethylene concentration levels. For ‘Golden Delicious’ and ‘Braeburn’, sesquiterpenes (alpha-farnesene) exhibited a fast response to ethylene emission followed by a rapid decline after the endogenous ethylene maximum peak. Carbonyl compounds displayed a different time evolution as compared to esters and terpenes and did not show any evident relationship with ethylene. Methanol and ethanol concentrations during the entire storage period did not change significantly. We show how multivariate analysis can efficiently handle the large datasets produced by PTR-ToF-MS and that the outcomes obtained are in agreement with the literature. The different volatile compounds could be simultaneously monitored with high time resolution, providing advantages over the more established techniques for the investigation of VOC dynamics in fruit post-harvest storage trials.     

Anthocyanin pattern of Tempranillo wines during ageing in oak barrels and storage in stainless-steel tanks

The anthocyanin fingerprint of Tempranillo wines made with grapes from two different vineyards of Rioja Alta has been studied by high-performance liquid chroatography during their ageing in oak barrels and their storage in stainless-steel tanks. The data were submitted to multifactorial analysis of variance, taking into account several factors: vineyard, age of wines, type of wine (free run wine and a coupage of free run wine and press wine), and type of container (oak barrels or stainless-steel tanks). The results indicate that both the length of ageing or storage and the vineyard where the grapes were grown affect the anthocyanin fingerprint of wines. The effect of the other two factors (type of wine and type of container) on the anthocyanin fingerprint of wines was quite low, despite the differences observed in several spectrophotometric parameters related to colour and phenolics.