On the synthesis of metadata tags for HTML files

RDFa, JSON-LD, Microdata, and Microformats allow to endow the data in HTML files with metadata tags that help software agents understand them. Unluckily, there are many HTML files that do not have any metadata tags, which has motivated many authors to work on proposals to synthesize them. But they have some problems: the authors either provide an overall picture of their designs without too many details on the techniques behind the scenes or focus on the techniques but do not describe the design of the software systems that support them; many of them cannot deal with data that are encoded using semistructured formats like forms, listings, or tables; and the few proposals that can work on tables can deal with horizontal listings only. In this article, we describe the design of a system that overcomes the previous limitations using a novel embedding approach that has proven to outperform four state-of-the-art techniques on a repository with randomly selected HTML files from 40 different sites. According to our experimental analysis, our proposal can achieve an F₁ score that outperforms the others by 10.14% ; this difference was confirmed to be statistically significant at the standard confidence level.     

Torii: An aspect-based sentiment analysis system that can mine conditions

Aspect-based sentiment analysis systems are a kind of text-mining systems that specialize in summarizing the sentiment that a collection of reviews convey regarding some aspects of an item. There are many cases in which users write their reviews using conditional sentences; in such cases, mining the conditions so that they can be analyzed is very important not to misinterpret the corresponding sentiment summaries. Unfortunately, current commercial systems or research systems neglect conditions; current frameworks and toolkits do not provide any components to mine them; furthermore, the proposals in the literature are insufficient because they are based on handcrafted patterns that fall short regarding recall or machine learning procedures that are tightly bound with a specific language and require too much configuration. In this article, we present Torii, which is a system that loads a collection of reviews, discovers the aspects on which they report, and summarizes the sentiment that is conveyed on them taking into account the existing conditions, if any. We also describe its architecture, our approach to mine conditions, and our experimental analysis on a large multilingual data set with reviews from multiple categories. To the best of our knowledge, Torii is the first proposal that addresses aspect-based sentiment analysis taking conditions into account.     

Detection of bodies in maritime rescue operations using unmanned aerial vehicles with multispectral cameras

In this study, we use unmanned aerial vehicles equipped with multispectral cameras to search for bodies in maritime rescue operations. A series of flights were performed in open‐water scenarios in the northwest of Spain, using a certified aquatic rescue dummy in dangerous areas and real people when the weather conditions allowed it. The multispectral images were aligned and used to train a convolutional neural network for body detection. An exhaustive evaluation was performed to assess the best combination of spectral channels for this task. Three approaches based on a MobileNet topology were evaluated, using (a) the full image, (b) a sliding window, and (c) a precise localization method. The first method classifies an input image as containing a body or not, the second uses a sliding window to yield a class for each subimage, and the third uses transposed convolutions returning a binary output in which the body pixels are marked. In all cases, the MobileNet architecture was modified by adding custom layers and preprocessing the input to align the multispectral camera channels. Evaluation shows that the proposed methods yield reliable results, obtaining the best classification performance when combining green, red‐edge, and near‐infrared channels. We conclude that the precise localization approach is the most suitable method, obtaining a similar accuracy as the sliding window but achieving a spatial localization close to 1 m. The presented system is about to be implemented for real maritime rescue operations carried out by Babcock Mission Critical Services Spain.     

VOCs and PAHs emissions from creosote-treated wood in a field storage area

In this study, the emissions of volatile organic compounds (VOCs, in this case aromatic hydrocarbons containing one benzene ring and furans) and polycyclic aromatic hydrocarbons (PAHs) from wood recently treated with creosote are examined. The VOCs and PAHs were identified and quantified in the gas phase. Additionally, the PAHs were quantified in the particulate phase. Glass multi-sorbent tubes (Carbotrap, Carbopack X, Carboxen-569) were used to hold the VOCs. The analysis was performed using automatic thermal desorption (ATD) coupled with capillary gas chromatography/mass spectrometry (GC/MS). PAHs vapours were collected on XAD-2 resin, and particulate matter was collected on glass fibre filters. The PAHs were analysed using GC/MS. The main components of the vapours released from the creosote-treated wood were naphthalene, toluene, m+p-xylene, ethylbenzene, o-xylene, isopropylbenzene, benzene and 2-methylnaphthalene. VOCs emission concentrations ranged from 35 mg m(-3) of air on the day of treatment to 5 mg m(-3) eight days later. PAHs emission concentrations ranged from 28 microg m(-3) of air on the day of treatment to 4 microg m(-3) eight days later. The air concentrations of PAHs in particulate matter were composed predominantly of benzo[b+j]fluoranthene, benzo[a]anthracene, chrysene, fluoranthene, benzo[e]pyrene and 1-methylnaphthalene. The emission concentrations of particulate polycyclic aromatic hydrocarbons varied between 0.2 and 43.5 ng m(-3). Finally, the emission factors of VOCs and PAHs were determined.     

Processing of ultrasonic array signals for characterizing defects. Part II: experimental work

This is Part II of the two-part paper aimed at integrating the numerical synthesis and experimental investigation of the ultrasonic wave propagation model for quantitative nondestructive evaluation. The first part of the paper focused on synthesizing and predicting measured signals using the boundary element method and the deconvolution technique based on the comparison between the signals obtained from defective and undamaged (reference) specimens. In the second part, we present an inversion technique which allows us to obtain the position and size of the defect. The inversion scheme is processing the frequency domain information rather than time-domain time-of-flights or vibration eigenmodes. This technique is tested experimentally for the case of a side-drilled hole with a non-trivial location in terms of standard pulse-echo techniques. It is shown that the scheme is particularly effective when the information of the defect is masked by other predominant signal components.     

An organic donor/acceptor lateral superlattice at the nanoscale

A precise control of the nanometer-scale morphology in systems containing mixtures of donor/acceptor molecules is a key factor to improve the efficiency of organic photovoltaic devices. Here we report on a scanning tunneling microscopy study of the first stages of growth of 2-[9-(1,3-dithiol-2-ylidene)anthracen-10(9H)-ylidene]-1,3-dithiole, as electron donor, and phenyl-C61-butyric acid methyl ester, as electron acceptor, on a Au(111) substrate under ultrahigh vacuum conditions. Due to differences in bonding strength with the substrate and different interactions with the Au(111) herringbone surface reconstruction, mixed thin films spontaneously segregate into a lateral superlattice of interdigitated nanoscale stripes with a characteristic width of about 10-20 nm, a morphology that has been predicted to optimize the efficiency of organic solar cells.     

Severe leukopenia and dysregulated erythropoiesis in SCID mice persistently infected with the parvovirus minute virus of mice

Parvovirus minute virus of mice strain i (MVMi) infects committed granulocyte-macrophage CFU and erythroid burst-forming unit (CFU-GM and BFU-E, respectively) and pluripotent (CFU-S) mouse hematopoietic progenitors in vitro. To study the effects of MVMi infection on mouse hemopoiesis in the absence of a specific immune response, adult SCID mice were inoculated by the natural intranasal route of infection and monitored for hematopoietic and viral multiplication parameters. Infected animals developed a very severe viral-dose-dependent leukopenia by 30 days postinfection (d.p.i.) that led to death within 100 days, even though the number of circulating platelets and erythrocytes remained unaltered throughout the disease. In the bone marrow of every lethally inoculated mouse, a deep suppression of CFU-GM and BFU-E clonogenic progenitors occurring during the 20- to 35-d.p.i. interval corresponded with the maximal MVMi production, as determined by the accumulation of virus DNA replicative intermediates and the yield of infectious virus. Viral productive infection was limited to a small subset of primitive cells expressing the major replicative viral antigen (NS-1 protein), the numbers of which declined with the disease. However, the infection induced a sharp and lasting unbalance of the marrow hemopoiesis, denoted by a marked depletion of granulomacrophagic cells (GR-1(+) and MAC-1(+)) concomitant with a twofold absolute increase in erythroid cells (TER-119(+)). A stimulated definitive erythropoiesis in the infected mice was further evidenced by a 12-fold increase per femur of recognizable proerythroblasts, a quantitative apoptosis confined to uninfected TER-119(+) cells, as well as by a 4-fold elevation in the number of circulating reticulocytes. Therefore, MVMi targets and suppresses primitive hemopoietic progenitors leading to a very severe leukopenia, but compensatory mechanisms are mounted specifically by the erythroid lineage that maintain an effective erythropoiesis. The results show that infection of SCID mice with the parvovirus MVMi causes a novel dysregulation of murine hemopoiesis in vivo.     

Seasonal and lactational changes in mineral composition of milk from Iberian red deer (Cervus elaphus hispanicus)

Milk minerals are important for calf growth, and they have other roles as well, such as immune regulation. This 2-yr study examined content of Ca, P, Mg, Na, K, Fe, and Zn in milk of 54 Iberian red deer hinds through 18 wk of lactation. Mean mineral composition of fresh milk was ash = 1.168 ± 0.007%, Ca = 2,330 ± 20 mg/kg, P = 640 ± 10 mg/kg, K = 1,100 ± 10 mg/kg, Na = 385 ± 3 mg/kg, Mg = 138 ± 1 mg/kg, Zn = 12.5 ± 0.2 mg/kg, and Fe = 0.65 ± 0.03 mg/kg. All minerals except Mg varied by week of lactation, but variation was usually <10% except for Fe (83% variation) and Zn (30% variation); both of those minerals increased as lactation proceeded. Increased concentrations of Fe and Zn in later lactation compensated for the reduction in milk production in mid and late lactation such that daily production was less variable for Fe (55% variation) or Zn (79% variation) than for other minerals (118 to 135% variation). Potassium content of milk decreased across time, but that effect occurred primarily during the last few weeks of lactation. Calving later vs. early in the calving season had variable effects on concentrations of different minerals: P, Mg, and K concentrations were not affected; Ca, Mg, and Na were all lower in milk from later calving hinds; and both Fe and Zn had higher concentrations in milk from hinds that calved later in the season. Lactating hinds seem to maintain a more stable daily yield of the microminerals Fe and Zn in milk compared with more variable concentrations of macrominerals as lactation progresses. Because of the essential role of Fe and Zn in immune function, a more stable supply of those minerals might be important to the health of growing red deer calves.