浅析淘宝网络刷单中的“水军”现象

本文阐述了网络刷客的产生与发展现状,探讨了网络刷客的法律规制问题。指出,对于网络刷客的规制,既要从道德层面去规制,更要尽快出台相关法律对网络刷客进行法律规制,一旦发现刷客从事侵害消费者权益的行为,就要给予严厉处罚。同时,要求第三方网络平台履行事先审查义务,严格把关进入网络平台的产品质量,从源头上保护消费者的权益。     

基于R软件和KNN算法的《红楼梦》作者分析

本文研究的是基于R软件和KNN算法的《红楼梦》作者分析,通过运用R软件将《红楼梦》前80回预定为第一类,后40回预定为第二类,进行KNN分类分析。最后得到的结论为:不拒绝《红楼梦》为两个作者所写。     

基于机器学习的P2P网络借贷违约风险识别模型比较——以“人人贷”为例

近年来,P2P网络借贷中借款人违约风险的识别问题被各互联网金融机构重点关注。文章以"人人贷"平台中的借贷数据为样本,运用CART决策树、支持向量机(SVM)、BP神经网络和随机森林等4种主流的机器学习算法,进行代价敏感学习,构建了4种违约风险识别模型并进行比较。研究发现,机器学习算法在多维度借贷数据下的违约风险识别中适用性较强,其中树模型的预测效果显著优于其他两种模型,尤以随机森林模型预测效果最佳。     

基于多类SVM武器装备预研项目风险预警模型及实证研究

在分析武器装备预研项目研究范畴及特点的基础上,进行全寿命周期内项目风险因素识别。对现有风险预警模型进行了比较分析,针对现有研究模型的不足,将支持向量机（Support Vector Machine, SVM）算法应用于武器装备预研项目风险预警,选取1-a-1方法对SVM进行多类扩展,选择RBF函数作为核函数,构建基于多类SVM的武器装备预研项目风险预警模型。最后,结合某军工研究所预研项目进行了实证研究,结果表明该方法拟合精度高、鲁棒性强,对于同类问题具有一定的推广价值。     

个人信用评估模型比较分析

文章以互联网金融机构客户信贷数据为基础,使用7种不同的数据挖掘方法建立个人信用评估模型,从预测准确率、模型外推性、第二类错误率、预期错误分类成本4个方面评价模型的综合信用评估能力。评价结果表明,使用分类树和K近邻分类算法建立的个人信用评估模型的综合信用评估能力最高;同时发现使用线性和非线性方法建立的模型各有特点,线性分类模型能够对违约客户进行有效识别,而非线性分类模型的预测精度较高。     

基于Sentinel-1A微波数据的土地利用信息提取方法对比

本文基于哨兵卫星Sentinel-1A双极化微波遥感数据,应用H/Alpha-Wishart、H/A/Alpha-Wishart两种非监督分类和Wishart迭代分类、支持向量机(SVM)两种监督分类方法分别对上海及长江入海口区域进行地物信息提取。对遥感数据进行预处理,利用H/A/Alpha极化目标分解提取极化特征参数;对影像进行非监督分类,对比非监督分类结果,计算分类精度;利用Wishart分类器和SVM分类器进行监督分类;参考Google earth等地图工具,对研究区的训练样本进行选取,将训练样本和极化参数带入分类器进行训练并不断调整参数,得到分类结果;进行精度分析,对比各分类方法的分类精度。由最终分类结果可知:监督分类对图像的解译效果最好,可解释性强,SVM监督分类结果最好,H/Alpha-Wishar最差,非监督分类方法中增加了A的H/A/Alpha-Wishart,非监督分类精度得到了提升。     

短期飞行冲突判定模型的研究

通过对航空器冲突判定模型的研究,阐述了两种模型的构建方式和具体应用方法,基于两种模型的建模方式和算法特点提出了一种新的冲突判定模型,新的模型简化飞行态势的分类和航空器保护区的设定,以航行差、航空器相对趋势及告警距离为主要影响因素,通过MATLAB利用雷达历史数据对新构建的模型进行了仿真验证,结果证明此模型在一定程度上提高了冲突探测的准确性,并满足实时性和虚警率的最低要求,保障了航空器的安全高效运行。     

采用分体式刷盒的电刷组件的焊接方法

本文阐述的是一种采用片状分体式刷盒的电刷组件的焊接方法,通过合适的焊接工装用焊料将两片分离的刷盒体与电刷焊接在一起,同时两片刷盒体也相互焊接在一起,其重点在焊接工装的设计。     

基于特征增量的新类识别方法研究

在分类应用的过程中,经常会出现新的类别,导致数据分布发生显著变化,使得原分类模型不再适用。如何识别新的类别使分类模型能适应其出现已经成为一个亟需解决的问题。本文提出基于特征增量的SVDD(支持向量数据描述)新类识别方法。该方法在SVDD算法的基础上,通过增加新特征,扩大特征空间维度从而提高模型对于新类的识别能力。在多个数据集上的实验结果表明,该方法能有效识别新类,使更新后的模型具有更高的准确度。     

银行定期存款客户预测——基于Lasso-Forest方法

从定性角度通过对数据进行描述性统计分析，给予银行营销策略建议。使用Lasso-Forest、Logistic、决策树及SVM模型分别进行建模分析，结合不平衡数据处理方法SMOTE算法，使用一系列评价指标进行模型效果评估，得到最优模型。所提出的Lasso-Forest组合模型具有比上述其他模型更精确的预测效果，识别准确率达到93%。使用提供的营销策略以及Lasso-Forest模型可以有效降低银行成本，对客户类型进行精确识别并加以针对性营销，以此达到增加银行定期存款储备的目的。     

On the Major Transformation of the Asia-Pacific Regional Structure北大核心CSSCISSCI

Recent years have witnessedsignificant changes in the Asia-Pacific regionalstructure featuring the shift in power balance among related countries as well as adjust-ments in their respective regional strategies. So what is the prevailing trend in thistransformation and what kind of regional order will likely emerge in the future？ To an-swer these questions, the paper tries to explore the following dimensions： the evolvingregional context, shift in power balance and power conversion, geo-political and geo-e-conomic developments and interactions between them, and evolving regional order. Asthe study shows, the broadening of regional geographical scope, the rise in the numberof actors involved, as well as changesin the relations among them, have combined tocast a pluralistic and complex flavor to the regional structure. While the power balanceamong China, the United States and Japan shifts, and power conversion is also occur-ring. China and Japan are paying increasing attention to their security roles, whereasthe United States under the Obama administration laid more stress on expanding itsgeo-economic clout. In spite of the intensifying geopolitical competition among China,the United States and Japan in recent years, a Sino-US strategic compromise-whichholds the key to regional geopolitics-isother front, some kind of geo-economicmore likely to take place in the future. On thecooperation will probably emerge. In the longerterm, it will be the geo-economic trend thatwill prevail over the geopolitical trend inthe region. Finally, the future evolution of regional order wiI1 mainly follow the eco-nomic logic, accompanied by the dilutionof its hegemonic and hierarchical nature.Hence, a pluralistic and complex regional community will emerge.     

The German Hanse and Bergen – new perspectives on an old subject

Abstract

The Hanseatic League and her kontors (trading posts) were an integral part of the medieval trade system in Northern Europe. Hanse merchants dominated many markets in this region and managed to maintain a monopoly-like role in towns like Novgorod and Bergen for many centuries. Consequently, one focus of historical research about the Hanseatic League lay on the mechanisms that enabled the Hanse merchants to keep that strong of a position for such a long time.

This article is part of the discussion about this topic. Based on Lübeck merchants with business interest in the Norwegian town of Bergen, the so-called Bergenfahrers (Bergen travellers), it will give new insights into the Hansards' economical and social organization. The three main points are (a) the integration of the Bergen market in the Hanseatic trade network as an intermediary trading place on the vivid East–West route of Hanse trade; (b) the importance of the Bergenfahrers within the Hanse merchants' economical, social and political networks; and (c) the structure of the Bergenfahrers network patterns in the late Middle Ages.

In addition to new insights into the important role of Bergen and the Bergenfahrers within Lübeck's and the other Wendish Hanse towns' trading system, this articles provides further proof for the usefulness of the methods of Social Network Analysis in medieval economic research.     

Organization and incentives in the age of sail

The British Navy in the age of sail was the most successful bureaucracy of its time. Its organization and incentive structures differed importantly from contemporaneous private sailing ventures, but closely resembled those of today’s large corporations. To induce efficient effort, the navy used a hierarchical tournament, in which sailors competed for higher pay that came with promotions based on relative performance. Promotion probabilities, the option value of future promotions, and the higher effort required of men in higher ranks and on larger vessels, combined to yield a highly skewed pay structure.     

Financial Reporting and Trade Credit: Evidence from Mandatory IFRS Adoption*

We investigate the effect of mandatory IFRS adoption on trade credit. We document that firms in countries that adopt IFRS receive more trade credit from their suppliers, consistent with improved financial reporting quality and comparability playing a role in facilitating informal financing. This increase is larger for countries with a low level of societal trust, a poor pre‐IFRS‐adoption information environment, and stronger legal enforcement. These cross‐sectional results suggest that the conditions under which higher‐quality information is made publicly available affect suppliers' decisions to provide trade credit. This increase is also larger for firms with greater exposure to foreign markets, a finding that highlights the importance of more comparable international financial reporting standards in facilitating cross‐country trade credit. We also find that IFRS adoption has a stronger positive effect on trade credit for firms with greater liquidity needs. Finally, we find that firms in countries that adopt IFRS also extend more trade credit to their customers. Overall, our results support the notion that financial reporting can have a causal effect on trade credit.     

区域工业产业竞争力指标体系的构建及其评价方法研究

文章通过建立省级区域工业产业竞争力评价指标体系,设计竞争力模型,利用SPSS统计软件,对各省工业产业竞争力进行了评价测定。     

中俄经济技术合作面临新机遇

中俄在十年内把各自的GDP翻一番的目标,为发展中俄经济技术合作提供了机遇.仅仅依靠商品贸易的增长很难大幅度提升中俄经贸关系,需要全面开展经济技术合作,特别是加强在自然资源开发、跨国投资、军工技术和产品贸易领域的合作.     

Penetration and Post-infection Development of Meloidogyne Incognita on Tobacco as Treated by Rhizobacteria Pseudomonas Fluorescens Strain P-72-10

The bio-control potential of rhizospere bacteria Pseudomonas fluorescens against plant-parasite nematode had been demonstrated. P. fluorescens had shown the effect to enhance tobacco resistance to root-knot nematode Meloidogyne incognita. Inoculation with P. fluoreseens in tobacco could lead to significant reductions in the number of juveniles that penetrated tobacco root and further life stage development of the juveniles. The number of juveniles penetrated into tobac- co root in treatment with P. fluorescens is significantly different from CK at 2DAI, 6DAI, 8DAI and 10DAI. Significant reduction and delayed development of juveniles that penetrated into tobacco root and treated were observed in treatment at 14DAI, 21DAI, 28DAI and 35DAI. In addition, P. fluoreseens treatment leads to a significant reduction in the number of eggs per egg-mass at 35 DAI. The results show P. fluorescens induced a continuously suppression on root-knot nema- tode M. incognita throughout their entire early infection phase of root penetration, subsequent life stage development and reproduction.