Search

Network embedding, as an approach to learning low-dimensional representations of nodes, has been proved extremely useful in many applications, e.g., node classification and link prediction. Unfortunately, existing network embedding models are vulnerable to random or adversarial perturbations, which may degrade the performance of network embedding when being applied to downstream tasks. To achieve robust network embedding, researchers introduce adversarial training to regularize the embedding learning process by training on a mixture of adversarial examples and original examples. However, existing methods generate adversarial examples heuristically, failing to guarantee the imperceptibility of generated adversarial examples, and thus limit the power of adversarial training. In this paper, we propose a novel method Identity-Preserving Adversarial Training (IPAT) for network embedding, which generates imperceptible adversarial examples with explicit identity-preserving regularization. We formalize such identity-preserving regularization as a multi-class classification problem where each node represents a class, and we encourage each adversarial example to be discriminated as the class of its original node. Extensive experimental results on real-world datasets demonstrate that our proposed IPAT method significantly improves the robustness of network embedding models and the generalization of the learned node representations on various downstream tasks. [ABSTRACT FROM AUTHOR]

This paper proposes a method combining blue the Haar wavelet and the least square to solve the multi-dimensional stochastic Itô-Volterra integral equation. This approach is to transform stochastic integral equations into a system of algebraic equations. Meanwhile, the error analysis is proven. Finally, the effectiveness of the approach is verified by two numerical examples. [ABSTRACT FROM AUTHOR]

Children are often perceived as beings who have very little or lack autonomy. The core of liberal education theory is to emphasize individual freedom and equality. In the context of liberal education, the purpose of education is to develop children's autonomy, which is in conflict with paternalistic intervention in education. This article defends that children have a degree of autonomy through an analysis of the concept and conditions of autonomy from a philosophical point of view. Furthermore, this paper demonstrates the compatibility of children's autonomy with adults' paternalistic intervention from the aspects of children's choice, respect for autonomy and future-oriented education. In the end, this paper proposes what is reasonable paternalism, that is, reasonable paternalism should be based on the moral equality of children and adults; reasonable paternalism should follow the principle of taking children as the center; reasonable paternalism is mainly achieved through the participation of children. [ABSTRACT FROM AUTHOR]

We investigated the hemodynamic parameters of pediatric PDA patients and focused on the influence of PDA size on pulmonary arterial pressure and the prevalence of pulmonary hypertension. A total of 52 patients aged between 2 months and 20 years who received transcatheter closure of a PDA from January 2018 to June 2022 in our institution were retrospectively recruited. Their hemodynamic parameters collected both by echocardiography and by cardiac catheterization were analyzed to delineate the influence of PDA size on the pulmonary vascular system. The echocardiographic-based ductal size and indexed PDA size were 1.93 mm (1.15–6 mm) and 4.05 mm/m2 (2.03–25.47 mm/m2), respectively. The pulmonary artery pressure measured was 20.83 mmHg (8–45 mmHg). We found a positive correlation between indexed PDA size and mean pulmonary arterial pressure (mPAP) (Pearson correlation coefficient = 0.47, p < 0.001). A subgroup analysis showed that 28 patients (53.8%) developed pulmonary hypertension (PH) (defined as mPAP > 20 mmHg). The median age of the PH group was 1.02 years [range: 0.19–8.64], which was significantly younger than the non-PH group's median age of 3.43 years [range: 0.42–19.96] (p = 0.001). The indexed PDA size for the PH group, 4.69 mm/m2, was significantly higher than that of the non-PH group, 3.2 mm/m2 (p = 0.004). The major risk factor for patients with PH was the PDA/BSA index, with an OR of 2.181 (95% CI, 1.224–3.887). Our demographic data showed younger patients with a higher PDA/BSA index are more likely to develop pulmonary hypertension. [ABSTRACT FROM AUTHOR]

microRNA-200c-3p (miR-200c-3p) has emerged as an important tumor growth regulator. However, its function in papillary thyroid carcinoma (PTC) is poorly understood. This study was conducted to investigate the role of miR-200c-3p in the progression of human PTC. The miR-200c-3p expression in human PTC tissues and cell lines was evaluated. The target relationship between miR-200c-3p and candidate genes was predicted through bioinformatic analysis and confirmed with a luciferase reporter assay. miRNA or gene expression was altered using transfection, and cell behavior was analyzed using CCK-8, wound healing, Transwell, and colony formation assays. The tumor-promoting effects of miR-200c-3p were evaluated by xenografting tumors with K1 cells in nude mice. The expression level of miR-200c-3p in human PTC tissues and cell lines markedly increased, and this increased expression was significantly associated with a worse overall survival. When inactivated, miR-200c-3p suppressed K1 cells' malignant behaviors, including decreasing proliferation and attenuating colony formation, migration, and invasion. Its inactivation also attenuated the development of xenografted K1 cells in nude mice. The effects of miR-200c-3p mimics on promoting the malignant behaviors of PTC cells were remarkably reversed by the overexpression of ATP2A2, as a downstream target of miR-200c-3p. miR-200c-3p acts as an oncogenic gene and promotes the malignant biological behaviors of human PTC cells, thereby directly targeting ATP2A2. This regulated axis may be used as a potential therapy of PTC. [ABSTRACT FROM AUTHOR]

Air pollution, especially heavy metals in PM2.5, has gradually become one of the most serious environmental problems in cities, which might cause various diseases. The trend of atmospheric concentrations of PM2.5-bound metals was analyzed, including Sb, As, Cd, Pb, Mn, and Ni in Chengguan District and Xigu District of Lanzhou (China) from 2015 to 2019. The optimal Autoregressive Integrated Moving Average (ARIMA) model was fitted with R 3.6.2 statistical software to predict the heavy metal concentrations in 2020. The concentrations of Sb, As, Cd, Pb, Mn, and Ni in both districts decreased to some extent in 2019 compared with the annual average in 2015, and the concentration of Pb in Chengguan District showed a decreasing trend year by year. The characteristics of the seasonal trend showed that Sb, As, Cd, and Pb were high in winter and low in summer and Mn increased in spring; Ni was high in summer and low in autumn in Chengguan District, while it was high in spring and low in autumn in Xigu District. There was a statistical significance (P < 0.05) in Mn concentrations between Chengguan District (non-industrial zone) and Xigu District. The ARIMA models, having good fitting properties and suitable for the prediction of heavy metal concentrations in two districts, have been established. Compared with 2019, the concentrations of As, Cd, and Mn in Chengguan District and Sb and Ni in Xigu District showed different degrees of increase by predicting with the optimal ARIMA model. Therefore, corresponding rectification measures should be taken to prevent and control the air pollution of Lanzhou City, reducing the occurrence of related diseases and improving the air quality. [ABSTRACT FROM AUTHOR]

A family of norm-based least squares support vector machines (LSSVMs) is the effective statistical learning tool and has attracted considerable attentions. However, there are two critical problems: (1) LSSVM is ill-conditioned or singular when the sample size is much less than feature number and thus causes over-fitting for small sample size (SSS) problem, while other norm-based LSSVMs own slower training speed and cannot deal with large scale data. (2) Norm-based LSSVMs pay less attention to the edge points that are important for learning the final classifier. To overcome the above drawbacks, by replacing L 2 - n o r m with L ∞ - n o r m in empirical loss, we construct a robust classifier, named L ∞ - L S S V M in short. After adjustment, L ∞ - L S S V M can detect edge points effectively and enhances the capability of the robustness and the generalization. Also, inspired by the idea of sequential minimal optimization (SMO), we design a novel SMO-typed iterative algorithm. The new algorithm not only guarantees the convergence of optimum solution in theory, but also owns the lower computational time and storage space when data size is large. Finally, extensive numerical experiments validate the above opinions again on four groups of artificial data, Non-i.i.d data: fault detection of railway turnout, four SSS datasets, two types of massive datasets and six benchmark datasets. [ABSTRACT FROM AUTHOR]

In common binary classification scenarios, learning algorithms assume the presence of both positive and negative examples. Unfortunately, in many practical areas, only limited labeled positive examples and large amounts of unlabeled examples are available, but there are no negative examples. In such cases, the algorithm that only exploits positive and unlabeled examples is needed. Such learning is termed as positive and unlabeled (PU) learning. In this paper, a novel classifier called global and local learning classifier (GLLC) for PU learning is proposed. The advantages of GLLC are as follows: (1) both intrinsic geometric structure and accurate positive information of PU data are exploited from global learning. (2) The smoothness and manifold of data are reflected sufficiently from local learning. (3) The algorithm of GLLC has faster training speed because the linear equations are solved. (4) The experiments on both synthetic and real datasets verify the above opinions and show that the classification result of GLLC is much better than those popular methods, such as LUHC, Pulce, BSVM, NB and so on. [ABSTRACT FROM AUTHOR]

Virus-induced gene silencing (VIGS) is a valuable reverse genetics tool to study gene function in a broad range of plants. As the efficiency of apple stable transformation is very low and it is still not easy to produce a large number of apple transgenic plants for most laboratories. Here, we developed an easy and effective transient transformation method of agro-inoculation for virus-induced gene silencing based on vacuum infiltration in genus of apples, the crabapple. The entire young tissue cultured crabapple plantlets were used as the plant material for infiltration, and a −90 kPa vacuum pressure and 19 °C day/18 °C night cultivation temperature were shown to result in highest VIGS efficiency. Infiltration with tobacco rattle virus containing a fragment of the transcription factor gene McMYB10, which regulates the synthesis of anthocyanin pigments, resulted in low levels of McMYB10 transcripts and characteristic faded red leaves. In addition, the expression of several anthocyanin biosynthetic genes was down-regulated in the transformed leaves. The results of this study indicate that the MYB10 gene represents a useful reporter gene for VIGS in crabapple and that the VIGS technique can be successfully applied to crabapple, thereby providing a platform for functional genetic studies in this and likely other Malus species. [ABSTRACT FROM AUTHOR]

Chalcopyrite dissolution was evaluated by bioleaching and electrochemical experiments with thermophile A. manzaensis ( Acidianus manzaensis) and mesophile L. ferriphilum ( Leptospirillum ferriphium) cultures at 65 °C and 40 °C, respectively. It was investigated that the bioleaching of chalcopyrite was stepwise. It was reduced to CuS at a lower redox potential locating in the whole bioleaching process by A. manzaensis at high temperature while only at initial days of bioleaching by L. ferriphilum at a relative low temperature. No reduced product was detected when the redox potential was beyond a high level (e.g., 550 mV (vs SCE)) bioleached by L. ferriphilum. Chalcopyrite bioleaching efficiency was substantially improved bioleached by A. manaensis compared to that by L. ferriphilum, which was mainly attributed to the reduction reaction occurring during bioleaching. The reductive intermediate CuS was more amenable to oxidation than chalcopyrite, causing enhanced copper extraction. [ABSTRACT FROM AUTHOR]

Bioleaching and electrochemical experiments were conducted to evaluate pyrrhotite dissolution in the presence of pure L. ferriphilum and mixed culture of L. ferriphilum and A. caldus. The results indicate that the pyrrhotite oxidation behavior is the preferential dissolution of iron accompanied with the massive formation of sulfur in the presence of L. ferriphilum, which significantly hinders the leaching efficiency. Comparatively, the leaching rate of pyrrhotite distinctly increases by 68% in the mixed culture of L. ferriphilum and A. caldus at the 3rd day. But, the accumulated ferric ions and high pH value produced by bioleaching process can give rise to the rapid formation of jarosite, which is the primary passivation film blocking continuous iron extraction during bioleaching by the mixed culture. The addition of A. caldus during leaching by L. ferriphilum can accelerate the oxidation rate of pyrrhotite, but not change the electrochemical oxidation mechanisms of pyrrhotite. XRD and SEM/EDS analyses as well as electrochemical study confirm the above conclusions. [ABSTRACT FROM AUTHOR]

This study aims to explore the prevalence and determinations of physical inactivity among hospital employees in Shanghai, China. A cross-sectional study of 4612 employees aged 19 to 68 years was conducted through stratified cluster sampling from different classes of Shanghai hospitals in 2011. The total physical activity was evaluated using the metabolic equivalent according to the Global Physical Activity Questionnaire. Among the participants, 38.5%, 32.3%, and 64.6% of the employees are inactive at work, commuting, and taking leisure time, respectively. Up to 41.8% of the men and 37.8% of the women ( P = 0.012) are physically inactive. When the age and educational level are adjusted, male doctors and medical technicians show a higher percentage of physical inactivity than male workers in logistics ( P = 0.001). Among females, employees who are working in second- and third-class hospitals show a higher proportion of physical inactivity than those who are working in community health care centers. Logistic regression analyses show that the odds ratios (ORs) of leisure-time physical inactivity associated with the intensity of physical activity at work are 2.259, 2.897, and 4.266 for men ( P < 0.001) and 2.456, 3.259, and 3.587 for women ( P < 0.001), respectively. The time during commuting activities is significantly associated with leisure-time physical inactivity in either sex (OR = 2.116 for men and 2.173 for women, P < 0.001). Hospital employees, particularly doctors and medical technicians, show a higher proportion of physical inactivity than other inhabitants in Shanghai. The time and intensity of activity at work and commuting are associated with leisure-time activities. [ABSTRACT FROM AUTHOR]

The bioleaching of chalcopyrite was investigated using a pure and mixed culture consisting of iron-oxidizing Leptospirillum ferriphilum ( L. ferriphilum) and sulfur-oxidizing Acidthiobacillus thiooxidans ( A. thiooxidans). The electrochemical tests were conducted to investigate the bioleaching behavior of chalcopyrite by various bacteria. Bioleaching efficiency of chalcopyrite in mixed culture is higher than that in the pure culture of L. ferriphilum alone. The iron-oxidizing L. ferriphilum plays a dominant role during bioleaching of chalcopyrite in the mixed culture of L. ferriphilum and A. thiooxidans. During bioleaching, certain values of redox potential are beneficial to the decomposition of chalcopyrite. Jarosite and sulfur are observed as products of bioleaching. The addition of A. thiooxidans during leaching by L. ferriphilum can change the electrochemical control steps of leaching. The corrosion current density is substantially promoted in the culture involving bacteria, especially in the mixed culture. [ABSTRACT FROM AUTHOR]

Characterization of the tracking error under continuous wave (CW) interference is the foundation of the interference detection and mitigation in receiver. It is helpful to select pseudo-random noise (PRN) codes with better code tracking performance under interference during the design of signal structure. Considering the effect of discrete spectral line of PRN code, we present a detailed analysis of tracking error of coherent delay locked loop (DLL) for the both situations that when the interference overlaps the spectral line and that when the interference is around the spectral line. In the first situation, the expression of the maximum tracking error for binary phase shift key (BPSK) modulated signal is obtained. The comparison of the tracking errors between finite and infinite frontend bandwidth cases in the first situation shows that the tracking error in the finite case is not like that in the infinite case and cannot be always reduced by narrowing the correlator spacing. The reason of this phenomenon is explained in detail. Further, we obtain the expression of the tracking error limitation as the correlator spacing tends to zero. It is more reasonable to take this limitation as the criterion of evaluating the code tracking performance for different PRN codes under interference. In the second situation, with the assumption of the infinite frontend bandwidth, we obtain the analytic expression of the tracking error. The analysis results show that the impact of the interference on tracking performance decays quickly when the interference leaves far away from the spectral line, and the diminution of correlator spacing and loop bandwidth both can reduce the tracking error efficiently. [ABSTRACT FROM AUTHOR]

Chalcopyrite oxidation rates were examined under various conditions in the presence of Leptospirillum ferriphilum, in which the effects of different pulp content, inoculation amount, external addition of Fe and initial pH value were studied. The bioleaching residues were investigated by X-ray diffractograms (XRD), scanning electron microscopy (SEM) and energy dispersion spectrum (EDS) analysis. The results show that low pulp concentration increases the leaching rate of copper, and external addition of Fe is also beneficial to leaching chalcopyrite. The changes of inoculation amount and initial pH from 1.6 to 2.5 have a little effect on the final leaching rate. The results also imply that Fe ions are important for bioleaching of chalcopyrite. At the end of bioleaching, jarosite and sulfur are observed on the surface of chalcopyrite residues by using XRD, SEM and EDS. With the passivation layer formed by jarosite and sulfur, the continuous copper extraction is effectively blocked. [ABSTRACT FROM AUTHOR]

At present, there are many factors affecting customer's choice when they decide on what to buy in a fuzzy environment. In the current customer's market, everyone wants things that are affordable but of high quality. Taking this into consideration, it becomes very important for each business to make a competent marketing strategic decision. They must be aware of the customer's thoughts to expand their market share and outdo their competitors. Clearly, the market still has a lot of factors to consider. Therefore, the critical issue is to identify with the customer's thoughts and construct a Customer's Choice Behavior Model (CCBM) for solving this problem of Customer's Choice. In this paper, a framework that considers customer's choice behavior, applies the graphtheorybased fuzzy Decision Making Trial and Evaluation Laboratory (Fuzzy DEMATEL) approach combined with Fuzzy Multiple Criteria Decision Making (FMCDM) is proposed. First, fuzzy DEMATEL is used to find the key factors and attributes in building the structure relations of an ideal Customer's Choice Behavior. Then, FMCDM is demonstrated to evaluate whether the firm's performance and customer's satisfaction have differences or not. Empirical studies of Lexus and BMW are illustrated to show how well Lexus' sales is doing, whether or not they had satisfied Customers' needs, and determine the magnitude of the gap between them in order to correct this with the right strategy. [ABSTRACT FROM AUTHOR]

The lack of structural information impeded the access of efficient luminescence for the exciplex type thermally activated delayed fluorescence (TADF). We report here the pump-probe Step-Scan Fourier transform infrared spectra of exciplex composed of a carbazole-based electron donor (CN-Cz2) and 1,3,5-triazine-based electron acceptor (PO-T2T) codeposited as the solid film that gives intermolecular charge transfer (CT), TADF, and record-high exciplex type cyan organic light emitting diodes (external quantum efficiency: 16%). The transient infrared spectral assignment to the CT state is unambiguous due to its distinction from the local excited state of either the donor or the acceptor chromophore. Importantly, a broad absorption band centered at ~2060 cm−1 was observed and assigned to a polaron-pair absorption. Time-resolved kinetics lead us to conclude that CT excited states relax to a ground-state intermediate with a time constant of ~3 µs, followed by a structural relaxation to the original CN-Cz2:PO-T2T configuration within ~14 µs. The development of exciplex-type hosts for thermally activated delayed fluorescence organic light-emitting diodes is hindered by a lack of structural information for these donor:acceptor blends. Here, the authors report the pump-probe Step-Scan Fourier transform IR spectra for a D:A exciplex host. [ABSTRACT FROM AUTHOR]