基于统计频率的文本分类特征选择算法研究*

通过分析χ2统计量（Chi-square, CHI）的缺陷和不足,针对它对低文档频的特征项不可靠,而且不能说明词条和类别的相关性的缺点,对其进行改进,提出统计频率(Statistical Frequency, SF )算法。实验结果表明,统计频率算法能够弥补这些不足,在文本分类中表现出良好的分类效果。     

基于概率神经网络的文本自动分类研究

本文提出了将一种径向基网络的重要变形———概率神经网络应用于文本自动分类的研究 ,与常用的K 最邻近法相比 ,具有一定的优势 ,特别是在训练集中各类的训练样本数很不平衡时 ;与BP等其他神经网络相比 ,其特点是需调节的参数少 ,不需确定隐层数、隐层中的神经元数量等网络结构 ,比较容易使用。此外 ,从研究中的不同特征选择的评价函数来看 ,它们对分类有一定的影响 ,应用X2 统计进行特征选择的分类正确率最高 ,其次是文本证据权 ,而期望交叉熵的效果最差 ,说明特征选择在文本自动分类中也是非常重要的。     

基于支持向量机的文本自动分类试验研究

提出了将支持向量机应用于文本自动分类的研究，与常用的K-最邻近法相比，无论是对训练数据集还是测试数据集均具有一定的优势，而且不同特征选择方法对支持向量机的影响要比K-最邻近法小。此外，从研究中的不同特征选择的评价函数来看，它们对分类有一定的影响，应用X2统计进行特征选择的分类正确率最高，其次是文本证据权，而期望交叉熵的效果最差，说明特征选择在文本自动分类中也是相当重要的。     

基于改进的SVM文本分类建模

SVM是一种新的分类工具,可是其核函数在数学上必须满足Mercer条件,使得具有良好全局分类性能的Sigmoid函数在SVM中应用受到限制。本文将Sigmoid核函数与云模型相结合,提出一种简单的核函数的实现方法。此方法不仅提高了SVM文本分类能力,而且明显地减少了平均的CPU执行时间。     

汉语个体量词的产生及其原因探讨

从历时的角度考察了汉语个体量词产生、发展和成熟时期的情况,和汉语个体量词产生的动因。认为汉语个体量词的产生是汉语范畴化和汉语类型特点制约的结果,同时反映了汉民族的认知视点。     

文本知识的自动分类方法初探

本文结合运用信息管理和人工智能的原理与技术，探讨了文本知识的自动分类方法，包括：自动归类与聚类方法、基于实例的学习分类方法和基于特征值的元学习方法。     

Extending WHIRL with background knowledge for improved text classification

Intelligent use of the many diverse forms of data available on the Internet requires new tools for managing and manipulating heterogeneous forms of information. This paper uses WHIRL, an extension of relational databases that can manipulate textual data using statistical similarity measures developed by the information retrieval community. We show that although WHIRL is designed for more general similarity-based reasoning tasks, it is competitive with mature systems designed explicitly for inductive classification. In particular, WHIRL is well suited for combining different sources of knowledge in the classification process. We show on a diverse set of tasks that the use of appropriate sets of unlabeled background knowledge often decreases error rates, particularly if the number of examples or the size of the strings in the training set is small. This is especially useful when labeling text is a labor-intensive job and when there is a large amount of information available about a particular problem on the World Wide Web.

现代日语量词“本”的认知考察

现代日语量词"本"的原型用法是表示细长物体的量,其扩张用法是表示运动时有轨迹的物体的量,边缘用法则是表示没有具体形状、但是内容上连贯,或时间上先后发生的、属于一个完整事件——比如人类的语言活动"记事"、"连载"、"讲演"等较新事物的量。另外日语中量词的使用,因不同事物,以及同一事物不同状态而不同。     

VC2OC2中宾语的语义类及范畴化现象

本文以对现代汉语语料库中的语料统计为基础,详细考察了不同的VC1OC2中宾语的语义类,发现不同的VC1OC2中宾语的语义类有相同之处,也有不同之处。文章指出,不同的VC1OC2中的宾语的语义类存在着范畴化现象。     

Leader prototypicality in sport: The implicit leadership theories of women and men entering sport management careers

Implicit leadership theories (ILTs) matter because they are used as the benchmark against which people determine who is a leader and who is not. This assessment informs their behavioral responses. People are thought to have a superordinate-level ILT representing their prototypical mental model of leadership and a series of basic-level ILTs. Each of these represents how people conceptualize leadership in particular domains (e.g., sport, politics, and popular culture). In this study, the authors generate the structures of basic-level sport management ILTs for women and men who are about to enter this industry. These form the baseline from which socialization to work, organizational cultures, and the sport management industry begins. The ILTs of women entering sport management careers have a 35-item, 6-factor structure (Sensitivity, Knowledgeable, Physical Attractiveness, Inspirational, Dedication, and Focused), whereas men entering the industry have a 32-item, 7-factor structure (Sensitivity, Dedication, Physical Attractiveness, Focused, Creativity, Inspirational, Courageous). The resulting sport management ILT profiles differ from superordinate ones by emphasizing physical and vitality attributes, incorporating emotional elements, and by eschewing antiprototypical elements. The paper ends with a discussion of the implications for leadership research in sport management.