结合TFIDF的Self-Attention-Based Bi-LSTM的垃圾短信识别

随着手机短信成为人们日常生活交往的重要手段,垃圾短信的识别具有重要的现实意义.针对此提出一种结合TFIDF的self-attention-based Bi-LSTM的神经网络模型.该模型首先将短信文本以词向量的方式输入到Bi-LSTM层,经过特征提取并结合TFIDF和self-attention层的信息聚焦获得最后的特征向量,最后将特征向量通过Softmax分类器进行分类得到短信文本分类结果.实验结果表明,结合TFIDF的self-attention-based Bi-LSTM模型相比于传统分类模型的短信文本识别准确率提高了2.1%–4.6%,运行时间减少了0.6 s–10.2 s.     

生姜多糖类物质的提取及抗氧化活性研究

目的以生姜为原料,采用超声波辅助法和热水浸提法提取生姜多糖类物质并进行抗氧化活性对比研究。方法经单因素试验结合响应面优化设计考察最优提取工艺参数,同时对2种方法提取的生姜多糖抗氧化活性进行对比分析。结果超声波辅助法提取生姜多糖最佳提取条件为:超声温度48℃,超声功率340W,超声时间21 min,液固比50:1(m L/g),在此条件下多糖得率为6.87%;热水浸提法的最佳提取条件为:温度72℃,时间164 min,液固比40:1(m L/g),在此条件下多糖得率为3.13%。抗氧化活性测定结果表明,超声波辅助法和热水浸提法提取生姜多糖的DPPH自由基清除能力的IC_(50)值分别为0.21 mg/m L和0.42 mg/m L,还原能力分别相当于维生素C的3.14%和0.5%,铁离子螯合能力的IC_(50)值分别为2.17 mg/m L和4.18 mg/m L,超声波辅助法提取的生姜多糖的DPPH自由基清除能力、还原力和金属螯合能力分别是热水浸提法提取的生姜多糖的2倍、6倍和1.9倍。结论生姜多糖具有一定的抗氧化活性,本研究可为生姜多糖的提取提供参考。     

耳廓点云形状特征匹配的路径跟随算法

路径跟随算法结合凸松弛方法与凹松弛方法,通过跟随凸凹问题的解路径,近似地求解图匹配问题,具有较高的匹配精度.将路径跟随算法用于耳廓特征图的匹配问题:首先,基于PCA方法构造耳廓点云的显著性关键点集合;然后,采用乘积型参数域上的单值二次曲面方法拟合关键点邻域内的点集,并将曲面的局部形状特征定义为耳廓的局部形状相似测度;第三,对关键点集合进行Delaunay三角剖分,得到关键点集合在三维空间内的拓扑结构图,并定义关键点图的整体结构差异测度;最后,记耳廓关键点图的组合差异测度为关键点图的整体结构差异测度与关键点上的局部形状相似测度的线性组合,并基于路径跟随算法快速求解关键点图之间的精确匹配.相关实验结果表明:与其他相关算法相比,该算法具有较高的匹配效率和匹配精度.     

SnO的歧化反应对SnTe热电性能的优化

PbTe基化合物是一种热电性能优良的中温区热电材料, 但铅的毒性限制了其广泛应用, 因此类似化合物SnTe引起了人们关注。但SnTe的载流子浓度较高和晶格热导率较大使其ZT值较低。本研究利用SnO歧化反应对SnTe热电性能实现了协同调控。热压烧结过程中SnO在500 ℃左右发生歧化反应生成Sn单质和单分散的SnO₂颗粒, Sn单质作为自掺杂可以填充SnTe中的Sn空位, 导致载流子浓度降低: 相比于SnTe基体, SnTe-6mol%SnO样品在600 ℃下的电阻率从6.5增大到10.5μΩ•m, Seebeck系数从105增大到146μV•K^-1。同时, 原位反应生成的SnO₂第二相单分散于晶界处, 多尺度散射声子传播而降低晶格热导率, SnTe-6mol%SnO样品晶格热导率在600 ℃下仅为0.6W•m^-1•K^-1, 相比于基体下降了33%左右, 从而使SnTe体系的热电性能得到明显提高。最终, 当SnO加入量为6mol%时, 样品在600 ℃下的ZT值~1, 相比于基体提升了一倍左右。     

Measuring quantum correlations using lossy photon-number-resolving detectors with saturation

The variance of difference of photocounts (VDPs) is an established measure of quantum correlations for quantum states of light. It enables us to discriminate between the classical correlation of a two-mode coherent state and the quantum correlation of a twin-beam state. We study the effect of loss and saturation of the photon number-resolving detector on the measurement of the VDPs. An analytic function is derived for this variance, both for the coherent and the twin-beam states. It is found that the VDPs is no longer a reliable entanglement measure in the nonlinear regime of the detector response but it remains useful in some range of values of average photon numbers of the incident light. We also quantify the linear regime of the detector with saturation which will be useful for calibration of the detector quantum efficiency.     

A Forecast Model of Hydrologic Single Element Medium and Long-Period Based on Rough Set Theory

On the basis of rough set theory, this paper presents the single element medium- and long-term classification forecast model that uses historical data of a hydrologic series as forecast factors. The minimal rule set, i.e., forecast pattern set, is achieved according to the principle of maximal attribute significance and rules frequency. Maximal support strength is put forward and applied to predict by using the model. The model is applied to forecast annual runoff of Dahuofang reservoir. The result indicates that the forecast model based on rough set can describe the relationship between forecast factors and forecast object efficiently and accurately. This model, which is composed of simple solution rules, can be easily understood and applied.     

Activation and pumping characteristics of Ti–Zr–V films deposited on narrow tubes

Non-evaporable getter (NEG) films are an inte-gral part of many particle accelerators. These films provide conductance-free evenly distributed pumping, a low th...     

基于SysML的机载软件分层精化建模与验证方法

机载软件被广泛应用于航空航天领域, 大幅提升了机载设备的性能.但随着机载软件规模逐渐增大、功能逐渐增多, 给软件的开发带来了难度, 如何保障机载软件的正确性和安全性也成为一个难题.基于模型的开发可以有效提升开发效率, 而形式化方法能够有效保障软件的正确性.为了降低开发难度, 同时保障机载软件的正确性、安全性, 本文提出一种基于SysML状态机图子集的机载软件分层精化建模与验证方法.首先使用SysML状态机图对机载软件的动态行为进行建模, 根据提出的精化规则, 对初始模型进行手动逐层精化得到精化设计模型.然后针对软件模型动态变化的特性, 将SysML状态机模型自动转换为时间自动机网络, 并从软件需求中手动提取形式化TCTL性质进行模型检验.其次, 为了实现编码自动化, 将SysML模型自动转换至Simulink, 利用Simulink Coder生成源代码.最后, 以一个自动飞行控制软件为例进行了开发和验证, 实验结果表明了该方法的有效性.     

3D Ear Shape Matching Using Joint α-Entropy

In this article, we investigate the use of joint α-entropy for 3D ear matching by incorporating the local shape feature of 3D ears into the jointα-entropy. First, we extract a su?cient number of key po...     

Developing a monitoring psychological stress index system via photoplethysmography

The purpose of this work was to assess the psychological stress index (PSI) by using oxyhemoglobin saturation by SpO₂ [i.e., the plethysmograph (PPG) signal], which can be measured easily and conveniently. We use the plethysmograph amplitude (PPGA) and the heart-beat interval (HBI) extracted from the SpO₂ waveform to obtain the stress index and to quantify it from 0 to 100. The respiration rate can also be extracted from the heart-rate interval, as shown in our previous research. Therefore, the PPG signal can display the heart rate, SpO₂ waveform, SpO₂ concentration, respiration rate, and psychological stress index. This methodology was tested on nine volunteer Taiwanese students under the stress of an English presentation. The experiment, which takes 30 min for each student, was separated into three stages: preparation, presentation, and discussion, and relaxation. The PSI values of these three stages were 49 ± 10, 60 ± 11, and 56 ± 10, respectively. The results were very successful, and showed PSI values changing from low, to high, and back to low during these three stages. In the near future, we hope to implement this system into a robotic wheelchair in order to monitor the PSI of elderly patients in a nursing home, and to evaluate their psychological condition based on this method.