基于WEKA平台的文本聚类研究与实现

文本聚类是文本挖掘领域的一个重要研究分支,是聚类方法在文本处理领域的应用。本文首先对基于空间向量模型的文本聚类过程做了较深入的讨论和总结。另外,本文回顾了现有的文本聚类算法,以及常用的文本聚类效果评价指标。在研究了已有成果的基础上,本文利用20Newsgroup文本语料库,针对向量空间表示模型,在开源的数据挖掘平台WEKA上实现了文本预处理和k-means聚类算法,并根据实际聚类效果,就文本表示、特征选择、特征降维等方面提出优化方案。     

大数据驱动的审计知识库建设与应用

国家审计的数据基础和审计方式随着大数据的发展而发生改变,其中文本分析技术逐渐深入应用到审计工作中。由于审计行业的领域专业性,有必要构建审计知识库以提高文本挖掘的准确性和可解释性。以大规模审计文本数据为驱动、以专家指导思想为基础、以自然语言处理为技术手段构建审计知识库,该知识库包括审计领域词典、领域文本训练语料与词向量模型、审计领域知识图谱,进而探讨审计知识库在审计工作中的具体应用。     

协整向量误设对相关检验的影响

常用于检验既定协整关系的统计量有tDF和tECM两种，但由于真实数据生成过程未知，估计模型中可能存在一定程度的协整向量误设，从而使统计量的分布特征受到影响。本文首先探讨tDF检验的隐含系数约束α=γ，即短期弹性等于先验长期弹性；其次分析零假设下两种统计量的分布特征，以及先验设定γ对信号噪声比q进而对tECM分布特征的影响；最后在局部备择假设下，给出两种统计量的渐近分布，并表明向量误设会降低协整检验的势，其程度与设定误差d正相关。     

结合IFC标准的建设项目中文文本分类研究

为促进项目参与方的合作和交流从而使项目更优质高效的完成,研究了结合IFC标准进行建设项目文档分类的方法。在对建设项目管理的特点进行深入分析的基础上,文章提出了将项目生命期中产生的大量的半结构化或非结构化的中文文本按照国际通用的IFC标准进行分类的方法,从而改进了文本的管理与利用效果。通过空间向量模型来表示中文文本,并采用夹角余弦的方法与国际通用的IFC标准中的实体进行相似度计算,最终实现中文文本的标准化分类,并通过案例分析验证了该方法的可行性。最后对本文提出的算法进行了评价,并提出了下一步的研究方向。     

Web文本挖掘及可视化

文本挖掘和文本可视化是计算机中重要应用技术,能够形象地高度概括文本信息中的核心内容,方便人们快速地理解和吸收文本中的核心思想。本文阐述文本挖掘预处理简要处理流程,然后阐述使用R软件进行文本挖掘,实现词项聚类、文本聚类、绘画词项云图、词项网络图等,找出其中隐藏的文本信息,并以可视化方式展现出来。最后对文本挖掘和文本可视化技术进行总结和展望。     

企业经营范围文本自动分类方法探究

随着各种数字化信息不断增长,如何对大量文档信息进行科学归类成为亟待解决的问题.文本自动分类方法成为目前解决该问题的一项关键技术.我国目前有超过1,000万的企业,企业经营范围是企业从事经营活动的具体描述.本文以企业经营范围数据为基础,根据其结构特征,以及与经济行业的关系,利用大规模文本数据的切分词优化、统计分类推断、属性关联分析等关键技术,通过在组织机构代码数据库中提炼相关数据进行实验比对分析,从而得出一种实用、高效的企业经营范围的文本自动分类方法.     

空间相关性的统计分析

空间自相关统计量是用于度量地理数据的一个基本性质,空间分析学者结合日益成熟的电脑科技GIS、空间计量方法、以及大型资料库,目的在精确地界定空间因素的重要性及影响力,空间权重矩阵用fij符号来表示空间的对象i,j的互相关联,fij=0就是表示空间权重矩阵的对角元素为零。空间权重矩阵有可以根据文中的几个函数方法来确定。     

信息不对称影响因素研究——以证券市场为例

本文通过引入带机制转换的向量自回归模型对信息交易概率模型进行改进,以描述信息不对称程度。通过面板回归的方法对中国证券市场信息不对称程度的可能影响因素进行了分析。实证结果显示,市场活跃度、市场预期和交易量对信息交易概率的影响比较大,市场微观结构和市场实际的信息状态对信息交易概率的影响比较小,但这四方面的解释因子对PIN都有不可替代的解释力,它们分别从不同的侧面解释了信息交易概率一半以上的信息。     

基于串频统计和词形匹配的分词系统

中文分词是中文文本挖掘的重要环节。中文分词的方法主要有基于辞典与规则和基于统计两种,"基于串频统计和词形匹配的分词系统"在一定程度上结合了中文分词两种方法的优点于一体,它充分利用文本本身的信息建立临时辞典,再配合一定的常用辞典,用词形匹配的方法达到分词的目的。     

大数据环境下基于文本挖掘的审计数据分析框架

大数据的浪潮推动着审计技术的变革,给审计模式和审计方法都带来了巨大的改变。传统的审计数据分析方法不能对半结构化以及非结构化数据进行分析,也无法满足大数据环境下审计信息化发展的要求,亟须提出新的审计数据分析思路和方法。在此背景下,文章提出了基于文本挖掘的审计数据分析框架,并阐述了采集与存储、挖掘与分析、总结与发布详细的审计数据分析流程。通过利用文本挖掘技术对采集的非结构化原始审计数据进行挖掘,根据明确的审计需求建立不同的文本挖掘模型,对审计数据进行分析,进而发现审计疑点,最终形成可理解的审计证据和审计线索。该框架的构建旨在为大数据审计提供新的思路,以降低大数据审计风险,提高审计质量。     

Incumplimiento del salario mínimo en Europa central y oriental

En este estudio se analiza el incumplimiento del salario mínimo durante el periodo 2003–2012 en diez países de Europa central y oriental con salario mínimo obligatorio nacional. A partir de los datos de EU‐SILC y la metodología propuesta por Bhorat, Kanbur y Mayet (2013), se cuantifica la incidencia y magnitud del incumplimiento. Se realizan análisis de regresión sobre determinantes individuales, laborales y macroeconómicos. Aunque la incidencia permanece relativamente baja, los trabajadores a los que se dirigen las políticas de salario mínimo parecen los más expuestos a su incumplimiento. Con el tiempo, una mayor incidencia se asocia a una razón más alta entre salario mínimo y salario medio.     

Cultural Economy Planning in Creative Cities: Discourse and Practice

While a growing body of research analyses the functional mechanisms of the cultural or creative economy, there has been little attention devoted to understanding how local governments translate this work into policy. Moreover, research in this vein focuses predominately on Richard Florida's creative class thesis rather than considering the wider body of work that may influence policy. This article seeks to develop a deeper understanding of how municipalities conceptualize and plan for the cultural economy through the lens of two cities held up as model ‘creative cities’ — Austin, Texas and Toronto, Ontario. The work pays particular attention to how the cities adopt and adapt leading theories, strategies and discourses of the cultural economy. While policy documents indicate that the cities embrace the creative city model, in practice agencies tend to adapt conventional economic development strategies for cultural economy activity and appropriate the language of the creative city for multiple purposes.     

Input/output ranges and performance: an examination of US machine tool producers

Despite the establishment of high-tech multinational enterprises (MNEs) in Ireland since the late 1950s, the country did not succeed in closing the technological gap with most of its EU counterparts. The weak National System of Innovation (NSI), and in particular low business and government R&D levels, combined with a lack of research and technology linkages between MNEs and indigenous firms, explain these poor results. However, the Shannon region in the West of the country presents some specific institutional characteristics that could theoretically make the region a strong technological enclave. The various actors in the region - i.e. the administrative, financial, political and research institutions, as well as the business organizations - have indeed infused a new type of economic development. After a brief review of the available concepts and models articulated around these actors and their interrelationships, the paper analyses whether the Shannon region, taken as an illustrative example, mirrors a specific economic and technological localized setting. The study is based on a survey of indigenous and foreign firms.     

Sixty years of discrimination and diversity research in human resource management: A review with suggestions for future research directions

This article reviews discrimination and diversity research published in Human Resource Management (HRM) over the past 60 years. While discrimination and diversity are very different constructs, it is often informative to study them together, because when people recognize each other's diversity, this can result in bias, stereotyping, and discrimination. We conducted bibliographic searches for terms related to discrimination and diversity as well as a manual search through every title and abstract published in HRM over the last 60 years to assess article relevance. The search resulted in 135 research articles with 136 unique studies (i.e., samples) which are reviewed in this article. Sex and race are the demographics that have been examined the most in HRM, while religion has been examined the least. Moreover, the number of studies examining lesbian, gay, bisexual, transgender, and queer (LGBTQ) employees in the workplace in HRM has grown quickly within the past 10 years, culminating in a recent meta‐analysis. Our review looks at some of the earliest research published, the most recent research published, and the overall trends we identified in the research over the years for both discrimination and diversity articles. We then make future research suggestions and recommendations to advance the study of discrimination and diversity in the coming years.     

Multinomial distributions applied to random sampling of particulate materials

When sampling a batch consisting of particulate material, the distribution of a sample estimator can be characterized using knowledge about the sample drawing process. With Bernoulli sampling, the number of particles in the sample is binomially distributed. Because this is rarely realized in practice, we propose a sampling design in which the possible samples have a nearly equal mass. Expected values and variances of the sample estimator are calculated. It is shown that the sample estimator becomes identical to the Horvitz–Thompson estimator in the case of a large batch-to-sample mass ratio and a large sample mass. Simulations and experiments were performed to test the theory. Simulations confirm that the round-off error due to the discrete nature of particles is negligible for large sample sizes. Sampling experiments were carried out with a mixture of PolyPropylene (PP) and PolyTetraFluorEthylene (PTFE) spheres suspended in a viscous medium. The measured and theoretical variations are in good agreement.     

The Early History of the Cumulants and the Gram-Charlier Series

The early history of the Gram-Charlier series is discussed from three points of view: (1) a generalization of Laplace's central limit theorem, (2) a least squares approximation to a continuous function by means of Chebyshev-Hermite polynomials, (3) a generalization of Gauss's normal distribution to a system of skew distributions. Thiele defined the cumulants in terms of the moments, first by a recursion formula and later by an expansion of the logarithm of the moment generating function. He devised a differential operator which adjusts any cumulant to a desired value. His little known 1899 paper in Danish on the properties of the cumulants is translated into English in the Appendix.     

Capturing the Implications of Services Trade Liberalization

This paper evaluates the impact of service sector trade liberalization on the world economy by a ten-region, eleven-sector CGE model with import embodied technology transfer from developed countries to developing countries. Simulation results show that service sector trade liberalization not only directly affects world service production and trade, but also has significant implications for other sectors in the economy. The major channel of the impact is through inter-industry input-output relations and TFP growth induced from services imported by developing countries from developed countries, which may be embodied with new information and advanced technology.     

Performance appraisal and culture: practice and attitudes in Hong Kong and Great Britain

This article compares the practice of and attitudes towards performance appraisal for managerial and professional staff in Hong Kong and Britain, and considers the extent to which actual practice and employee preferences are in alignment. Findings suggest that appraisal may be more widespread in Hong Kong than in Britain. However, British appraisal tends to be more participative and to place greater emphasis on discussing objectives, development and career plans. Hong Kong appraisals appear to be more directive and Hong Kong respondents perceive a higher level of 'negative' appraiser behaviour. In spite of this, Hong Kong respondents show if anything slightly more confidence in the utility of appraisal than do British respondents. They show stronger support for appraisal's use for reward and punishment and less support for the objectives-setting and training and development uses than do the British sample. Hong Kong respondents are more likely than their British counterparts to favour involving a more senior manager in appraisal, and they are less likely to prefer more frequent appraisals. There was little evidence that Hong Kong respondents had a stronger preference for group-based appraisal criteria, although they did show more support than the British sample for the use of personality as a basis for appraisal. Overall, the suggestion is that appraisal has been adopted in Hong Kong organizations but that the practice of appraisal has been adapted to suit the cultural characteristics of the society.     

Narrative construction of expatriate experience and career cycle: discursive patterns in Finnish stories of international

Expatriate experience is not only a disconnected occasion for cross-cultural anxiety and adjustment but also an important event in the process of self-development and learning. Following this view and arguing for a discursive approach, the paper focuses on ways in which expatriates themselves tell and interpret their development and movement across expatriate career cycle. Meaning systems connecting expatriate job with previous and following work experiences in career stories of Finnish engineers and managers were identified using a combination of narrative and discourse analysis. No evidence was found of an autonomous expatriate discourse but, in contrast, expatriate career cycle was narrated using available organizational repertoires of development and career. Describing and discussing the meaning-making properties of three identified discourses - bureaucratic, occupational and enterprising - the paper emphasizes the organizational environment of expatriate experience while acknowledging the limits of these contemporary career vocabularies in addressing individual learning and change in cross-cultural settings.     

Internet professionals: job skills for an on-line age

The paper explores the increasing popularity of the Internet technology Singapore. It argues that the diffusion of Internet technology to South-East Asia would create new duties and responsibilities for human resource (HR) managers. One such duty is the identification of the skills which professional Internet staff would need to perform their work effectively in organizations. Using a triangulation research method, empirical study has identified the new professional Internet jobs which have emerged the IT industry in Singapore and the skills that professional Internet personnel require do their jobs effectively. Analysis of the findings indicates that professional Internet skills involve both technical and creative abilities and that, in explaining or defining professional Internet skills, it is necessary to use both the technical and the social conceptions of skill. This argument is based on the view that the creative skills include tacit skills so that a definition based solely on the technical conception of skill cannot capture the various dimensions of Internet skills. The implications of the research findings for human resource managers are raised.