ACTSSD: social spammer detection based on active learning and co-training

The rumors, advertisements and malicious links are spread in social networks by social spammers, which affect users’ normal access to social networks and cause security problems. Most methods aim to detect social spammers by various features, such as content features, behavior features and relationship graph features, which rely on a large-scale labeled data. However, labeled data are lacking for training in real world, and manual annotating is time-consuming and labor-intensive. To solve this problem, we propose a novel method which combines active learning algorithm with co-training algorithm to make full use of unlabeled data. In co-training, user features are divided into two views without overlap. Classifiers are trained iteratively with labeled instances and the most confident unlabeled instances with pseudo-labels. In active learning, the most representative and uncertain instances are selected and annotated with real labels to extend labeled dataset. Experimental results on the Twitter and Apontador datasets show that our method can effectively detect social spammers in the case of limited labeled data.

     

Greenhouse gases mitigation by CO<Subscript>2</Subscript> reforming of methane to hydrogen-rich syngas using praseodymium oxide supported cobalt catalyst

This study focuses on the potential of hydrogen-rich syngas production by CO₂ reforming of methane over Co/Pr₂O₃ catalyst. The Co/Pr₂O₃ catalyst was synthesized via wet-impregnation method and characterized for physicochemical properties by TGA, XRD, BET, H₂-TPR, FESEM, EDX, and FTIR. The CO₂ reforming of methane over the as-synthesized catalyst was studied in a tubular stainless steel fixed-bed reactor at feed ratio ranged 0.1–1.0, temperature ranged 923–1023 K, and gas hourly space velocity (GHSV) of 30,000 h^?1 under atmospheric pressure condition. The catalyst activity studies showed that the increase in the reaction temperature from 923 to 1023 K and feed ratio from 0.1 to 1.0 resulted in a corresponding increase in the reactant’s conversion and the product’s yields. At 1023 K and feed ratio of 1.0, the activity of the Co/Pr₂O₃ catalyst climaxed with CH₄ and CO₂ conversions of 41.49 and 42.36 %. Moreover, the catalyst activity at 1023 K and feed ratio of 1.0 resulted in the production of H₂ and CO yields of 40.7 and 40.90 %, respectively. The syngas produced was estimated to have H₂:CO ratio of 0.995, making it suitable as chemical building blocks for the production of oxygenated fuel and other value-added chemicals. The used Co/Pr₂O₃ catalyst which was characterized by TPO, XRD, and SEM-EDX show some evidence of carbon formation and deposition on its surface.     

纳米Fe_2O_3作为润滑油添加剂的摩擦磨损性能

采用烧结方法制备得到纳米α-Fe_2O_3,利用X射线衍射仪(XRD)和透射电镜(TEM)对样品进行了表征,研究了纳米α-Fe_2O_3作为液体石蜡添加剂的摩擦磨损性能。结果表明,添加纳米添加剂后,润滑油摩擦系数改变不明显,但磨损率显著降低。磨损率与纳米α-Fe_2O_3添加剂的表面积和添加量密切相关,当纳米材料的表面积为47m~2/g时磨损率最低,磨损率值降低为3.78×10~(-15) m~2/g;添加量为1.0wt%,润滑油磨损率最低。     

数控机床操控器人机界面优化设计研究

目的控制器和显示器共同组成数控机床控制面板的整体界面,是机床接受与执行命令的交互平台。从人机交互角度,聚焦于控制器的界面布局进行优化设计。方法基于人机工程学理论的优化流程,采用现场访谈、现场观察等方法,将人机工程学的理论转化为具体考核指标。结论从人机工程学的研究角度出发,结合工业设计、用户体验设计学等内容,从功能区域分区、操纵器色彩设计等方面优化数控机床控制器的界面,提高系统整体效能及使用者满意度,为数控机床系统开发设计提供参考。     

Composite bending-dominated hollow nanolattices: A stiff,cyclable mechanical metamaterial

Manufacturing ultralight and mechanical reliable materials has been a long-time challenge. Ceramic-based mechanical metamaterials provide significant opportunities to reverse their brittle nature and unstable mechanical properties and have great potential as strong, ultralight, and ultrastiff materials. However, the failure of ceramics nanolattice and degradation of strength/modulus with decreasing density are caused by buckling of the struts and failure of the nodes within the nanolattices, especially during cyclic loading. Here, we explore a new class of 3D ceramic-based metamaterials with a high strength–density ratio, stiffness, recoverability, cyclability, and optimal scaling factor. Deformation mode of the fabricated nanolattices has been engineered through the unique material design and architecture tailoring. Bending-dominated hollow nanolattice (B-H-Lattice) structure is employed to take advantages of its flexibility, while a few nanometers of carbonized mussel-inspired bio-polymer (C-PDA) is coherently deposited on ceramics’ nanolayer to enable non-buckling struts and bendable nodes during deformation, resulting in reliable mechanical properties and outperforming the current bending-dominated lattices (B-Lattices) and carbon-based cellulose materials. Meanwhile, the structure has comparable stiffness to stretching-dominated lattices (S-Lattices) while with better cyclability and reliability. The B-H-Lattices exhibit high specific stiffness (>10⁶?Pa·kg^?1·m^?3), low-density (～30?kg/m³), buckling-free recovery at 55% strain, and stable cyclic loading behavior under up to 15% strain. As one of the B-Lattices, the modulus scaling factor reaches 1.27, which is lowest among current B-Lattices. This study suggests that non-buckling behavior and reliable nodes are the key factors that contribute to the outstanding mechanical performance of nanolattice materials. A new concept of engineering the internal deformation behavior of mechanical metamaterial is provided to optimize their mechanical properties in real service conditions.     

Probabilistic calibration of discrete element simulations using the sequential quasi-Monte Carlo filter

The calibration of discrete element method (DEM) simulations is typically accomplished in a trial-and-error manner. It generally lacks objectivity and is filled with uncertainties. To deal with these issues, the sequential quasi-Monte Carlo (SQMC) filter is employed as a novel approach to calibrating the DEM models of granular materials. Within the sequential Bayesian framework, the posterior probability density functions (PDFs) of micromechanical parameters, conditioned to the experimentally obtained stress–strain behavior of granular soils, are approximated by independent model trajectories. In this work, two different contact laws are employed in DEM simulations and a granular soil specimen is modeled as polydisperse packing using various numbers of spherical grains. Knowing the evolution of physical states of the material, the proposed probabilistic calibration method can recursively update the posterior PDFs in a five-dimensional parameter space based on the Bayes’ rule. Both the identified parameters and posterior PDFs are analyzed to understand the effect of grain configuration and loading conditions. Numerical predictions using parameter sets with the highest posterior probabilities agree well with the experimental results. The advantage of the SQMC filter lies in the estimation of posterior PDFs, from which the robustness of the selected contact laws, the uncertainties of the micromechanical parameters and their interactions are all analyzed. The micro–macro correlations, which are byproducts of the probabilistic calibration, are extracted to provide insights into the multiscale mechanics of dense granular materials.     

A facile synthesis of supported amorphous iron oxide with high performance for Cr(Ⅵ) removal from aqueous solution under visible light irradiation

A series of supported iron oxide nanoparticles were prepared by impregnation with Fe(NO_3)_3 supported on TiO_2,followed by low-temperature calcination. Scanning electron microscopy(SEM), X-ray diffraction(XRD), X-ray photoelectron spectroscopy(XPS), UV–vis diffuse reflectance spectra and BET have been used to characterize the samples. These iron oxide-impregnated TiO_2 were examined for photocatalytic reduction of Cr(Ⅵ). The experiments demonstrated that Cr(Ⅵ) in aqueous solution was more efficiently reduced using Fe_2O_3/TiO_2 heterogeneous photocatalysts than either pure Fe_2O_3 or TiO_2 under visible light irradiation. All TiO_2 supported samples were somewhat active for visible light photoreduction. With an optimal mole ratio of 0.05-Fe/Ti, the highest rate of Cr(Ⅵ) reduction was achieved under the experimental conditions. We also compared the photoreactivity of TiO_2 supported iron oxide samples with that supported on Al_2O_3 and ZrO_2. It can be noted that iron oxide nanoparticles deposited on high surface area supports to increase the solid-liquid contact area renders it considerably more active. Noticeably,iron oxide cluster size and dispersion are important parameters in synthesizing active, supported Iron oxide nanoparticles. In addition, the interaction between iron oxide and TiO_2 was proposed as the source of photoactivity for Cr(Ⅵ) reduction.     

Entrapping multifunctional dendritic nanoparticles into a hydrogel for local therapeutic delivery and synergetic immunochemotherapy

Nano Research - Developing multifunctional nanoparticles to support new therapy models is a promising and challenging task to address the current dilemma on antitumor treatment. Herein, we...     

Neural Machine Translation by Fusing Key Information of Text

When the Transformer proposed by Google in 2017, it was first used for machine translation tasks and achieved the state of the art at that time. Although the current neural machine translation model can generate high quality translation results, there are still mistranslations and omissions in the translation of key information of long sentences. On the other hand, the most important part in traditional translation tasks is the translation of key information. In the translation results, as long as the key information is translated accurately and completely, even if other parts of the results are translated incorrect, the final translation results’ quality can still be guaranteed. In order to solve the problem of mistranslation and missed translation effectively, and improve the accuracy and completeness of long sentence translation in machine translation, this paper proposes a key information fused neural machine translation model based on Transformer. The model proposed in this paper extracts the keywords of the source language text separately as the input of the encoder. After the same encoding as the source language text, it is fused with the output of the source language text encoded by the encoder, then the key information is processed and input into the decoder. With incorporating keyword information from the source language sentence, the model’s performance in the task of translating long sentences is very reliable. In order to verify the effectiveness of the method of fusion of key information proposed in this paper, a series of experiments were carried out on the verification set. The experimental results show that the Bilingual Evaluation Understudy (BLEU) score of the model proposed in this paper on the Workshop on Machine Translation (WMT) 2017 test dataset is higher than the BLEU score of Transformer proposed by Google on the WMT2017 test dataset. The experimental results show the advantages of the model proposed in this paper.