基于信任链的骚扰电话预判技术研究

当前,骚扰电话乱象屡禁不止,严重影响了人民群众的正常生活,甚至威胁到个人财产安全,因此简单有效地识别骚扰电话技术成为亟待解决的问题。区别于现有的对电话号码进行黑白名单标记、大数据分析等识别方法,提出基于信任链的骚扰电话预判技术,从可信度量的角度计算来电号码的可靠程度,并针对高危受害群体的典型应用场景进行分析,通过对骚扰电话的预判,降低其受到电话骚扰甚至电话诈骗的可能性。     

Thermodynamic assessment of the Ga–Se–Te system

In this study, the Ga–Te binary system was reassessed by means of the CALPHAD method using a modified lattice stability parameter for Te as well as experimental data for this binary system. The two-sublattice ionic solution model was applied for the liquid phase, and the intermediate phases were described by the sublattice model. A set of self-consistent thermodynamic parameters was optimized for all the phases in the Ga–Te binary system, which reproduced the phase diagram and the thermodynamic properties well. Using the reevaluated Ga–Te system, previously assessed Ga–Se system, and modified Se–Te system, a critical evaluation of the Ga–Se–Te ternary system was performed. The calculated vertical sections, isothermal sections, and liquidus projection agreed reasonably well with the experimental data. Immiscibility in the liquid phase was observed, and the origin of this behavior is discussed from a thermodynamic perspective.     

卡尔费休法测定液态氯乙烯中水含量的研究

采用卡尔费休法测定了液态氯乙烯单体中的水含量。试验确定了采样工具、样品的预处理方式及选用溶剂的最优方案,并进行了精密度与准确度的验证。结果表明卡尔费休法测定液态氯乙烯单体中的水含量准确、快速,可以用来准确、及时地指导生产。     

Effect of laver powder on textual,rheological properties and water distribution of squid (Dosidicus gigas) surimi gel

In order to ameliorate the gel quality of Dosidicus gigas surimi, the effects of laver powder on gel properties, rheological properties, and water-holding capacity (WHC) were investigated. Results indicated that the addition of laver powder could significantly increase the hardness, chewiness, and breaking force of surimi gels. However, the texture indexes and gel strength began to decline when additional amount exceeded 0.6%. Rheological results demonstrated that the addition of laver powder increased the storage modulus (G′) and viscosity of surimi, prolonged protein denaturation temperature in surimi gels. Moreover, the WHC of surimi gel was improved with the increase of laver powder. Further analyses in low-field nuclear magnetic resonance revealed that laver powder could shorten the transverse relaxation time, enhanced the combination with water, and altered the distribution of different water categories. The proportion of bound water and immobilized water reached its maximum and minimum at 0.6% of laver powder, respectively. Correlation analyses showed that WHC of surimi gel was negatively correlated well with the proportion of loose-bound water, but positively correlated with the strong-bound water and free water. In conclusion, the results supported that 0.6% was the optimal additional amount of laver powder for the squid-based surimi production based on the current ingredients of surimi products.     

Combined application of electrolysed water and ultrasound to improve the sanitation of knives in the meat industry

This study investigated the effect of the combination of basic electrolysed water (BEW) and slightly acid electrolysed water (SAEW) with ultrasound (US) for cleaning and sanitation of the knives used in slaughtering processes. The knives were sonicated in a US bath using two modes of operation (normal and sweep) in two steps: (i) 5 min with BEW and (ii) 10 min with SAEW at 35 °C. The microbiological counts and the possible changes in the physical structure of the knives were evaluated. The association BEW + SAEW + US, in the sweep mode, provided lower mesophilic, enterobacterial, Staphylococcus aureus and yeast counts when compared to the values recommended by the international legislation. In addition, these conditions removed all organic residues from the knife blades and promoted the highest migration rate of the residues to the US water bath (12.35 mg/L·min), without modifications in the knife blades. Thus, cleaning and sanitation of knives was feasible with the association of BEW + SAEW + US, which could be a useful alternative for the meat industry.     

公民水素养评价表征因素模型构建与分析

针对我国水资源紧缺、水生态损害、水环境污染等相关水问题,以提升全民爱水惜水节水护水的意识和能力为目标,从水素养概念出发,通过扎根理论质性研究方法对公民水素养评价的表征因素进行筛选和提炼,最终得出包含水知识、水态度、水行为为主范畴的10个表征因素并运用解释结构模型(ISM)深入剖析各因素之间的内在联系。在此基础上建立了3级多层次递阶结构模型,进一步精确描述系统各要素两两之间的直接关系。结果表明:公民水素养评价的表征因素自上而下形成了表象层、中间层和根源层等3个层级,可直观、清晰地反映表征因素之间的递阶、层级关系;水态度在各表征因素中具有调节作用以及各表征因素之间存在内在逻辑关系。研究成果为提高公民水素养水平提供一定的科学依据和政策参考。     

薄膜铂电阻元件稳定性测试与分析

高端军事应用环境对测温传感器提出了小结构、高精度、良好的长期稳定性等要求,采用溅射方式制备的薄膜铂电阻元件成为了首选。通过分析薄膜铂电阻元件的结构、工艺,探讨影响稳定性的主要因素,包括薄膜缺陷、杂质、电迁移、结构封装应用与温度应力等。选择薄膜铂电阻在水三相点下的电阻值作为稳定性的比较基准,设计了一款不大于3mK的高精度测试系统。通过对薄膜铂电阻元件进行高温电寿命试验和温度冲击试验,评估得出薄膜铂电阻元件长期稳定性达到10mK,满足设计需求。     

面向云计算应用层演化的隐私保护方法研究

云计算应用层中的组合服务具有演化属性,因此,隐私数据在服务组合过程中,用户的隐私数据可能会因为服务或服务流程的演化而暴露。根据服务演化的特征,以描述逻辑为基础,提出了一种面向云计算应用层演化的隐私保护方法。对隐私协议进行形式化描述;根据服务的演化特征,对服务的演化进行监控,保证满足用户的隐私需求;利用实例研究证明该方法的正确性与可行性。     

基于多尺度的贝叶斯模型显著性检测

针对传统基于贝叶斯模型的显著性检测算法存在准确率不理想的问题,提出了一种基于多尺度的贝叶斯模型显著性检测算法。通过超像素分割算法(SLIC)将原图分割成不同尺度的超像素,根据超像素边界信息得到背景种子,进而通过距离计算和多尺度融合得到背景先验;对原图进行颜色增强,采用Harris算子对增强图进行检测角点求得凸包,融合不同尺度下的超像素得到凸包先验;融合背景先验和凸包先验得到最终先验;利用颜色直方图和凸包计算似然概率;将最终先验和似然概率通过贝叶斯模型计算显著图。在公开数据集MSRA1000、ECSSD上与多种传统算法进行准确率和召回率对比,该算法有更好的表现。     

基于主动学习的离群点集成挖掘方法研究

离群点检测任务通常缺少可用的标注数据，且离群数据只占整个数据集的很小一部分，相较于其他的数据挖掘任务，离群点检测的难度较大，尚没有单一的算法适合于所有的场景。因此，结合多样性模型集成和主动学习思想，提出了一种基于主动学习的离群点集成检测方法OMAL（Outlier Mining based on Active Learning）。在主动学习框架指导下，根据各种基学习器的对比分析，选择了基于统计的、基于相似性的、基于子空间划分的三个无监督模型作为基学习器。将各基学习器评判的处于离群和正常边界的数据整合后呈现给人类专家进行标注，以最大化人类专家反馈的信息量；从标注的数据集和各基学习器投票产生的数据集中抽样，基于GBM（Gradient BoostingMachine）训练一个有监督二元分类模型，并将该模型应用于全数据集，得出最终的挖掘结果。实验表明，提出方法的AUC有了较为明显的提升，且具有良好的运行效率，具备较好的实用价值。