基于近似权重计算的网络安全威胁评估方法

传统的AHP方法由于判断值受人为影响因素太大,导致在衡量多因素权重时常常出现结果不一致,影响结论的准确性和评估结果的可信性。本文以网络风险评估中的威胁评估为例,建立基于“威胁对资产的影响一安全属性一攻击”的威胁评估模型,利用层次化的计算模型评估威胁的风险等级,从给出的区间判断矩阵入手,将区间判断矩阵一致逼近到一般的数字判断矩阵,得到各层元素的近似权重。通过实例验证,该方法能精确地、自动量化实时风险势态状况。     

多壁碳纳米管基质固相分散萃取-HPLC法测定蔬菜中的啶虫脒残留

建立了以多壁碳纳米管（MWCNTs）为萃取剂的基质固相萃取（MSPD）、HPLC测定蔬菜中啶虫脒残留的方法。将样品与工业级碳纳米管、无水硫酸镁混合研磨后装柱,并用乙酸乙酯洗脱,高效液相色谱法测定。采用C8柱（3.0mm×100mm,1.8μm）分离,甲醇-水（3∶2,v/v）为流动相,流速0.10mL/min,紫外检测波长250nm,进样量2.5μL,柱温为30℃。结果表明,该方法对蔬菜样品中啶虫脒的回收率为84.0%¹06.0%,精密度为0.1%¹2.5%,方法最低检出限为0.0085mg/L。该法简便、快速、准确。      

运用增广矩阵束方法稀布优化平面阵

基于增广矩阵束方法(Matrix Enhancement and Matrix Pencil,MEMP),以使用尽可能少的阵元逼近期望的方向图为目标,提出了一种求解阵元位置和设计激励幅度的新方法.首先对期望平面阵的方向图进行采样得到离散的数据集,再构造增广矩阵,对此增广矩阵进行奇异值分解(Singular Value Decomposition,SVD),确定逼近期望方向图所需的最小阵元数目;基于广义特征值分解求解两组特征值,并根据类基于旋转不变技术的信号参数估计(Estimating Signal Parameters Via Rotational Invariance Techniques,ESPRIT）对这两组特值配对;在最小二乘准则下求解稀布面阵的阵元位置和激励.仿真试验验证了该方法在稀布平面阵优化问题中的高效性和数值精度.     

一种基于社交网络社区的组推荐框架

面向用户群组的推荐主要面临如何有意义地对群组进行定义并识别,以及向群组内用户进行有效推荐两大问题。该文针对已有研究在用户群组定义解释性不强等存在的问题,提出一种基于社交网络社区的组推荐框架。该框架利用社交网络结构信息发现重叠网络社区结构作为用户群组,具有较强的可解释性,并根据用户与群组间的隶属度制定了考虑用户对群组贡献与用户从群组获利的4种聚合与分配策略,以完成组推荐任务。通过在公开数据集上与已有方法的对比实验,验证了该框架在组推荐方面的有效性和准确性。     

多导体传输线瞬态响应研究

随着多导体传输线内各导体之间间距的减小, 导体之间的近邻效应对传输线的分布参数和传输特性的影响越来越大.为此, 我们针对三种典型的传输线结构, 分别建立了基于矢势有限元方法分析的多导体传输线的模型, 并分析了近邻效应对磁通密度和分布电感的影响.利用提出的方法计算了同轴传输线的单位长度分布电感, 并将它与采用解析方法得到的结果进行比较来证明该方法的正确性.计算双线传输线在不同间距时的单位长度电感, 与理论分析得到的结果相比较验证了导线间距越小, 近邻效应对单位长度电感的影响越大.最后, 计算考虑了近邻效应的耦合微带线的电感矩阵, 并将它与其他不考虑近邻效应的方法得到的结果相比较, 说明近邻效应对传输线电感矩阵的影响.     

一种基于MapReduce的知识聚类与统计机制

网络文献知识库中的海量资源及其分类的粗粒度,导致学习者容易在文献检索和阅读过程出现认知迷航和知识过载问题。该文提出一种基于MapReduce的知识聚类与统计机制：首先,提出基于MapReduce的共现矩阵构建算法MR-CoMatrix;其次,将共现矩阵与相似度系数结合构建相似度矩阵;然后,通过Z Scores对相似度矩阵进行标准化;最后,使用离差平方和法(Ward,s method)对相似度矩阵进行聚类,生成树状的知识聚类谱系图;基于聚类结果,提出基于MapReduce的知识文献统计算法MR-Statistics,对每个分类的知识属性进行统计。实验结果表明：将MR-CoMatrix和MR-Statistics方法应用于网络文献知识库进行知识聚类和统计,达到较理想的聚类精度和计算效率,实现了细粒度知识聚类和多维统计,同时减少了时间开销。     

Endogenous Modulation of Extracellular Matrix Collagen during Scar Formation after Myocardial Infarction

Myocardial infarction is remains the leading cause of death in developed countries. Recent data show that the composition of the extracellular matrix might differ despite similar heart function and infarction sizes. Because collagen is the main component of the extracellular matrix, we hypothesized that changes in inflammatory cell recruitment influence the synthesis of different collagen subtypes in myofibroblasts, thus changing the composition of the scar. We found that neutrophils sustain the proliferation of fibroblasts, remodeling, differentiation, migration and inflammation, predominantly by IL-1 and PPARγ pathways (n = 3). They also significantly inhibit the mRNA expression of fibrillar collagen, maintaining a reduced stiffness in isolated myofibroblasts (n = 4–5). Reducing the neutrophil infiltration in CCR1^−/− resulted in increased mRNA expression of collagen 11, moderate expression of collagen 19 and low expression of collagen 13 and 26 in the scar 4 weeks post infarction compared with other groups (n = 3). Mononuclear cells increased the synthesis of all collagen subtypes and upregulated the NF-kB, angiotensin II and PPARδ pathways (n = 3). They increased the synthesis of collagen subtypes 1, 3, 5, 16 and 23 but reduced the expression of collagens 5 and 16 (n = 3). CCR2^−/− scar tissue showed higher levels of collagen 13 (n = 3), in association with a significant reduction in stiffness (n = 4–5). Upregulation of the inflammation-related genes in myofibroblasts mostly modulated the fibrillar collagen subtypes, with less effect on the FACIT, network-forming and globular subtypes (n = 3). The upregulation of proliferation and differentiation genes in myofibroblasts seemed to be associated only with the fibrillar collagen subtype, whereas angiogenesis-related genes are associated with fibrillar, network-forming and multiplexin subtypes. In conclusion, although we intend for our findings to deepen the understanding of the mechanism of healing after myocardial infarction and scar formation, the process of collagen synthesis is highly complex, and further intensive investigation is needed to put together all the missing puzzle pieces in this still incipient knowledge process.     

Genetic Deletion of Mmp9 Does Not Reduce Airway Inflammation and Structural Lung Damage in Mice with Cystic Fibrosis-like Lung Disease

Elevated levels of matrix metalloprotease 9 (MMP-9) and neutrophil elastase (NE) are associated with bronchiectasis and lung function decline in patients with cystic fibrosis (CF). MMP-9 is a potent extracellular matrix-degrading enzyme which is activated by NE and has been implicated in structural lung damage in CF. However, the role of MMP-9 in the in vivo pathogenesis of CF lung disease is not well understood. Therefore, we used β-epithelial Na⁺ channel-overexpressing transgenic (βENaC-Tg) mice as a model of CF-like lung disease and determined the effect of genetic deletion of Mmp9 (Mmp9^-/-) on key aspects of the pulmonary phenotype. We found that MMP-9 levels were elevated in the lungs of βENaC-Tg mice compared with wild-type littermates. Deletion of Mmp9 had no effect on spontaneous mortality, inflammatory markers in bronchoalveolar lavage, goblet cell metaplasia, mucus hypersecretion and emphysema-like structural lung damage, while it partially reduced mucus obstruction in βENaC-Tg mice. Further, lack of Mmp9 had no effect on increased inspiratory capacity and increased lung compliance in βENaC-Tg mice, whereas both lung function parameters were improved with genetic deletion of NE. We conclude that MMP-9 does not play a major role in the in vivo pathogenesis of CF-like lung disease in mice.     

Indirect Immobilised Jagged-1 Enhances Matrisome Proteins Associated with Osteogenic Differentiation of Human Dental Pulp Stem Cells: A Proteomic Study

The indirect immobilisation of Jagged-1 (Jagged-1) promoted osteogenic differentiation of human dental pulp cells (hDPs). Furthermore, the analysis of the Reactome pathway of RNA sequencing data indicates the upregulated genes involved with the extracellular matrix (ECM). Hence, our objective was to investigate the effects of Jagged-1 on proteomic profiles of human dental pulp stem cells (hDPSC). hDPSCs were cultured on the surface coated with human IgG Fc fragment (hFc) and the surface coated with rhJagged1/Fc recombinant protein-coated surface. Cells were differentiated to the osteogenic lineage using an osteogenic differentiation medium (OM) for 14 days, and cells cultured in a growth medium were used as a control. The protein component of the cultured cells was extracted into the cytosol, membrane, nucleus, and cytoskeletal compartment. Subsequently, the proteomic analysis was performed using liquid chromatography–tandem mass spectrometry (LC-MS). Metascape gene list analysis reported that Jagged-1 stimulated the expression of the membrane trafficking protein (DOP1B), which can indirectly improve osteogenic differentiation. hDPSCs cultured on Jagged-1 surface under OM condition expressed COL27A1, MXRA5, COL7A1, and MMP16, which played an important role in osteogenic differentiation. Furthermore, common matrisome proteins of all cellular components were related to osteogenesis/osteogenic differentiation. Additionally, the gene ontology categorised by the biological process of cytosol, membrane, and cytoskeleton compartments was associated with the biomineralisation process. The gene ontology of different culture conditions in each cellular component showed several unique gene ontologies. Remarkably, the Jagged-1_OM culture condition showed the biological process related to odontogenesis in the membrane compartment. In conclusion, the Jagged-1 induces osteogenic differentiation could, mainly through the regulation of protein in the membrane compartment.