面向语料库标注的汉语依存体系的探讨

实现大规模真实文本的处理, 是信息化社会的迫切要求, 也是国际计算语言学界的一个战略目标目前一项迫在眉睫的任务是建立一套满足大规模真实文本处理的语言处理体系, 包括分词的标准、词的分类体系、句法体系和语义体系。其中句法体系是核心环节。本文提出并论证了依存语法是合乎大规模真实文本处理要求的句法体系, 并结合汉语的特点, 研究了汉语的依存语法, 划分了种依存关系。最后简要讨论了依存语法的一些应用     

Justinian lexicography

This paper describes two research projects, both involving Latin and Greek lexicography. They are undertaken at the University of Florence and at the Italian National Research Council respectively. The one involves the creation of a Dictionary of Justinian's constitutions based on the emperor's legislative lexicon formed in the Corpus Iuris and elsewhere. The most demanding aspect of this task has been the creation of the Dictionary of the Novellae. The other project involves the creation of a Lexicon of the Novellae in the Authenticum version. Anna Maria Bartoletti Colombo is a Research Director of the CNR and teaches at the University of Florence. She took part in the first experiments in computational linguistics with the Historical Dictionary of the Italian language of the Accademia della Crusca. Proceeding to the Italian Legal Dictionary of the CNR, she concentrated her interests in the language of law. Her major works are Vocabolario delle Costituzioni Greche e Latine di Giustiniano and Lessico delle Novallae di Giustiniano as described in her article in this issue.     

Aphrodisiacs from plant and animal sources—A review of current scientific literature

The use of aphrodisiacs dates back thousands of years in Chinese, Indian, Egyptian, Roman, and Greek cultures. Although the scientific basis of these substances was not understood, aphrodisiacs were valued for their ability to enhance the sexual experience. Their use allowed for human procreation and the ability to obtain a sexually fulfilling relationship. Aphrodisiacs used historically include ambrein, Bufo toad, Spanish fly, yohimbine, Tribulus terrestris, horny goat weed, muira puama, MACA root, Panax ginseng, nutmeg, saffron, and cacao. Previous studies on these substances have shown potential aphrodisiac properties using animal models and in human clinical trials. Aphrodisiacs were shown to relax corpus cavernosum smooth muscle tissue in animals, improve erection quality in humans and animals, or increase sexual behavior and satisfaction in humans and animals. Although most studies showed positive effects of aphrodisiacs on sexual enhancement, more studies are needed to understand their mechanism of action. The need for clinical trials using larger populations is also evident to prove the effectiveness of aphrodisiacs for human use. This paper will review recent scientific studies conducted on these commonly used aphrodisiacs, and determine whether the results support or refute their use for human sexual enhancement.     

一种规则与统计相结合的汉语分词方法

汉语自动分词是中文信息处理领域的一项基础性课题,对现有的汉语分词方法作了简单的概述和分析,然后提出了一种新的分词方法,该方法基于一个标注好了的语料库,并且结合了规则和语料库统计两种分词方法。     

基于Ontology的语料库系统

针对传统语料库的缺陷,提出了基于本体的语料库,并设计分析了这个具有语义的语料库系统.系统不仅能够建立和维护基于本体的语料库,并且使得建立的语料库具有动态的、开放的、基于语义的等多种特点.对系统的架构作出了较为详细的分析,并设计了系统的框架,最后对系统作了较为客观的评价.系统不仅具有广阔的应用前景,也具有重大的理论价值.     

Online and off-line visualization of meeting information and meeting support

In current meeting research we see modest attempts to visualize the information that has been obtained by either capturing and, probably more importantly, by interpreting the activities that take place during a meeting. The meetings being considered take place in smart meeting rooms. Cameras, microphones and other sensors capture meeting activities. Captured information can be stored and retrieved. Captured information can also be manipulated and in turn displayed on different media. We survey our research in this area, look at issues that deal with turn-taking and gaze behavior of meeting participants, influence and talkativeness, and virtual embodied representations of meeting participants. We stress that this information is interesting not only for real-time meeting support, but also for remote participants and off-line consultation of meeting information.     

基于基频包络的音节聚类研究

对汉语TTS系统的大规模语料库做了基本的韵律参数统计，分析了音节的韵律特征与其所在的韵律结构位置以及韵律结构边界的关系。进一步，对有调音节样本集基于基频包络采用k中心点算法进行聚类，通过听辩实验检验了聚类结果。并分析了音节聚类与其所在韵律结构之间的对应关系。     

文本切分知识获取及其应用

文章描述了一种从熟语料中自动获取文本切分知识的机器学习的方法。该方法从已切分标注加工的熟语料中自动获取错误的文本切分形式及其相应的上下文信息,以及正确的切分形式,形成知识库,并将获取的知识再应用到新语料的加工过程,从而进一步提高机器切分的正确率;同时,采用滚动式的方法,建造具有较高加工质量的大规模语料库。     

一种模仿人类的自动文本分类算法

1.引言 Internet上有着大量的且快速增长的文本,文本是信息和知识的宝贵资源。随着Internet的快速发展,不久的将来,人们所需要的大部分信息都可以在网上找到。Internet正在成为人类的信息宝库,但是随着网上信息的爆炸性增长,人们想从这个信息宝库中获得自己所需要的信息已经变得日益困难,因此,如何快速有效地获得有用的信息已成为人们十分关     

汉语词性标注方法的研究

1 引言自然语言中,表达意义的符号(词)往往在各个层面上有歧义。在句法层面上,一个词可以兼好几种词性;在语义层面上,一个词可能有多个义项。词性歧义是由语言中的兼类词,即具有不止一个词性特征的词所引起的,只有在一定的上下文语境关系中,词所表现