面向中文关系抽取的句子结构获取方法

在关系抽取中,神经网络模型是目前最常用的技术之一,然而现有神经网络模型很少考虑句子中两个实体之间的结构特征。该文针对关系抽取任务的特点,提出了基于神经网络模型的句子结构获取方法。该方法通过对关系实例中两个实体的位置进行特殊标记,使神经网络模型能够有效捕获句子中关于实体的结构信息。为了验证方法的有效性,分别采用两种主流的神经网络模型进行对比实验,实验结果表明,该方法在ACE 2005中文关系抽取数据集上的抽取性能得到显著提升,超出对比工作约11个百分点,表明该方法能有效提升关系抽取任务的性能。     

The impact of CCT on driving safety in the normal and accident situation: A VR-based experimental study

Correlated color temperature (CCT) of the light source inside the road tunnel plays a crucial role in ensuring driving safety, which is demonstrated by previous studies that CCT influences not only visual effects but also non-visual effects. Although conventional laboratory experiments could simulate the CCT environment inside the tunnel to some extent, they fail to restore driving experience, let alone simulate driving behavior in the accident situation. It has largely remained unclear whether and how CCT would influence visual/non-visual performance of the subjects who are performing driving tasks, especially in the accident situation. Motivated by this gap, a virtual-reality-based framework for assessing the influence of CCT on the visual and non/visual performance in normal driving situation and in accident situation was proposed. In this study, tunnel models under seven different CCTs were created and a rear-end accident was designed in the tunnel. By integrating analog driving equipment, all participants were required to perform virtual driving tasks both in the normal situation and in the accident situation. The non-visual performance (driving fatigue) and the visual performance (reaction time) of the participants were collected and analyzed. Results show that the CCT of light source inside the tunnel was significant on the driving fatigue of the driver who was performing driving task, and it also had a significant impact on the visual performance when the driver was faced with a rear-end accident. Detailed experimental methodology, behavioral explanations underlying these findings, validity of results and practical implications are also discussed in the paper.     

Difficulty and complexity definitions for assembly task allocation and assignment in human–robot collaborations: A review

This paper presents a literature review on the different aspects of task allocation and assignment problems in human–robot collaboration (HRC) tasks in industrial assembly environments. In future advanced industrial environments, robots and humans are expected to share the same workspace and collaborate to efficiently achieve shared goals. Difficulty- and complexity-aware HRC assembly is necessary for human-centric manufacturing, which is a goal of Industry 5.0. Therefore, the objective of this study is to clarify the definitions of difficulty and complexity used to encourage effective collaboration between humans and robots to leverage the adaptability of humans and the autonomy of robots. To achieve this goal, a systematic review of the following relevant databases for computer science was performed: IEEE Xplore, ScienceDirect, SpringerLink, ACM Digital Library, and ASME Digital Collection. The results extracted from 74 peer-reviewed research articles published until July 2022 were summarized and categorized into four taxonomies for 145 difficulty and complexity definitions from the perspectives of (1) definition-use objectives, (2) evaluation objectives, (3) evaluation factors, and (4) evaluation variables. Next, existing definitions were primarily classified according to the following two criteria to identify potential future studies on the formulation of new definitions for human-centric manufacturing: (1) agent specificity and (2) common aspects in manual and robotic assemblies.     

融合空洞卷积的轻量化目标检测

为了轻量化模型,便于移动端设备的嵌入,对YOLOv4网络进行了改进.首先,用MobileNetV3作为主干网络,并使用深度可分离卷积替换加强特征提取网络的普通卷积,降低模型参数量;其次,在104×104特征图输出时融合空洞率为2的空洞卷积,与52×52的特征层进行特征融合,获取更多的语义信息和位置信息,细化特征提取能力,提升模型对极小目标的检测性能;最后,将原来的池化层使用3个5×5的Maxpool进行串联,减少计算量,提升检测速度.实验结果表明,在华为云2020数据集上,改进算法的mAP比YM算法提高了2.33%,在公共数据集VOC07+12上, mAP提高了3.12%, FPS比原来的YOLOv4算法提高了一倍多,参数量降低至原来的18%,证明了改进算法的有效性.     

轴承故障诊断中特征选取技术

针对轴承故障诊断建模中如何通过筛选有效特征提高模型诊断准确率的问题,提出一种新的特征选取方法。在计算所得特征集合中,利用诊断模型直接对特征进行判断,将高于阈值的诊断准确率对应的特征(组合)选取为显著特征,以显著特征导向选取方式,找到候选特征集合中维度低、诊断准确率高的特征。试验结果表明,本研究提出的方法可筛选出有效特征,降低模型参数、减少样本需求量、提高模型准确率,提升了故障诊断的效率。     

A secure rational quantum state sharing protocol

A novel rational protocol to share two arbitrary qubits among multiple parties is investigated in this paper. First, the protocol is presented, which is learned from Li et al.’s protocol. Second, the utility, security, correctness, fairness, Nash equilibrium, and Pareto optimality of our scheme are discussed in detail, where the utility, correctness, and fairness of rational quantum state sharing protocols are creatively given because the agent who recovers the state plays a different and more important role. Another important point is that assumptions about our protocol are more practical and suitable than existing protocols.     

Optimum design of a wideband planar N-way fork power divider with arbitrary power division and input-to-output impedance matching

An effective design procedure based on method of least squares is proposed for multi-section and multi-output fork power divider/combiner with arbitrary power division ratios among its outputs in a specified frequency bandwidth together with impedance matching among its arbitrary source/load impedances. Transmission and scattering matrices are obtained for its equivalent circuit. An error function is constructed based on design specifications on its output power division ratios, isolations among output ports, return losses at its input and output ports and source/load impedances in a desired frequency bandwidth. The design procedure is fully developed, which determines the widths and lengths of microstrip line sections and resistor values. A design example is described for unequal power division ratio and unequal input/output impedances. A 3-section and 3-output fork power divider is designed for equal power division and load/source impedances for 2–12 GHz band. It is fabricated and measured. Variations of its transmission coefficients are less than 0.5 dB, isolations at its outputs are better than −15 dB and reflection coefficients at its ports are better than −10 dB. Excellent agreement is observed among the results of the proposed design procedure, full wave computer software and measurement data.     

Effect of coexistence of H2S on production of hydrogen and solid carbon by methane decomposition using Fe catalyst

Biogas derived from livestock manure and food residue contains CO₂ and H₂S as well as methane. The effect of CO₂ and H₂S coexistence on the production of hydrogen and solid carbon by methane decomposition over iron oxide catalysts was investigated. The catalytic activity for methane decomposition was decreased by the coexistence of H₂S. Moreover, the activity decrease was aggravated by the coexistence of CO₂ as well as H₂S, and higher temperature was required to mitigate the activity decrease by the coexistence of CO₂. By increasing the amount of catalyst, the upstream catalyst was preferentially poisoned, but the downstream catalyst developed catalytic activity thanks to its sacrifice. With 2 g of catalyst, the maximum conversion of pure methane was about 85% at 840 °C, but it was slightly less than 80% in the presence of H₂S or H₂S + CO₂. When the catalyst amount was increased to 4 g, the conversion of pure methane was about 90% at 800 °C, but 84% in the presence of H₂S and 80% in the presence of H₂S + CO₂. The poisoning by H₂S was irreversible at low temperatures but became reversible at higher temperatures. Since H₂S is adsorbed by the deposited carbon, the procedure for further removal of H₂S may be omitted. The coexistence of H₂S also affected the shape of the deposited carbon. Although carbon-based catalysts are known to be effective for methane decomposition in the presence of H₂S, iron oxide catalysts have the advantage of superior methane conversion at low temperatures. By flowing methane with CO₂ and H₂S from the downstream side after the reaction flowing from the upstream side for a certain period of time, the catalytic lifetime was drastically extended and the amount of hydrogen and solid carbon produced was dramatically increased, compared to the case of flowing from upstream all the way.     

Directional Modulation Based on a Quantum Genetic Algorithm for a Multiple-Reflection Model

Directional modulation is one of the hot topics in data security researches. To fulfill the requirements of communication security in wireless environment with multiple paths, this study takes into account the factors of reflections and antenna radiation pattern for directional modulation. Unlike other previous works, a novel multiple-reflection model, which is more realistic and complex than simplified two-ray reflection models, is proposed based on two reflectors. Another focus is a quantum genetic algorithm applied to optimize antenna excitation in a phased directional modulation antenna array. The quantum approach has strengths in convergence speed and the globe searching ability for the complicated model with the large-size antenna array and multiple paths. From this, a phased directional modulation transmission system can be optimized as regards communication safety and improve performance based on the constraint of the pattern of the antenna array. Our work can spur applications of the quantum evolutionary algorithm in directional modulation technology, which is also studied.