让Word校对得更准确一些

大家知道,Word中提供了文字校对功能,当我们完成内容录入工作之后,可以用它来检查文字错误。但是,这样的校对往往不能令人满意,原因是Word只是用系统默认的词库来进行校对,系统词库包含的词量有限,并不能按用户自身的语言习惯来进行校对,因而还有许多错误校不出来。如果能将我们自己在日常使用输入法的过程中自然形成的个性词库或者网络上提供的最新流行词库移植过来,就能大大增强Word的校对功能,让它来帮我们识别出更多的错误。     

手写文字识别机的机械结构设计

现在计算机的速度不断提高,但还需要采用穿孔纸带或卡片输入信息。通常一个穿孔员每天最大穿孔量为24000冲程。在处理大量数据时,需要大量穿孔校对人员和穿孔校对设备及大面积的工作间。而手写文字识别机只需要操作员对手写文字用肉眼校对便可输入,这样随时都可处理任意量的信息,而且当急于上机或     

各种校对软件之我见

在进行书刊报章和上网信息的编撰工作中，大量文稿输入计算机后的校对工作是一件繁琐而易出统漏的工作，通常是由校对人员面对计算机屏幕或打印件上的文字同原稿校校，由于人工操作难免一时疏忽，因此往往造成经过再三校核也有错误出现。在一些全国知名的计算机专业报刊中经常可见一些常用字的错误，诸如“跨掉的企业”、“深表谦意”之类似是而非的别字（计算机输入汉字不可能出现错字，而只会出现别字），使报刊的形象大打折扣。有些错误的出现并非校对者的疏忽，而可能是校对者本身文化功底较为薄弱造成，他们将一些别字当成正确的词语而…     

一种基于Google的OCR结果校对新方法

为了提高OCR识别率,许多校对算法被提出用作后处理.这些方法利用自然语言规律统计大量的语料库进行语法判断.由于语料库规模有限且无法及时更新,导致一些网络新词、专用名词等常被错误处理.为此,文中将传统的语料库和Google知识库相结合,利用Google知识库获得网络新词等出现的频率,建立N-Gram模型,利用词语间的接续关系进行查错,再借助Google的拼写校对功能和词语之间的可信度进行改错.相比传统的方法,该方法的语料库源于互联网,对一些新词有更好的校对效果,更适合图像中嵌入的文字信息识别结果的校正.     

文档文字录入智能批阅的实现

正确率是衡量计算机操作人员文字录入水平的重要指标.如何对录入的文本进行智能化批阅从而获得正确率?就此问题进行深入的研究与分析,并在此基础上用.NET编制了校对程序,完全达到智能校对的效果.列出所有程序代码,还对程序的设计思路、蝙制方法等进行了详尽的阐述.     

基于CSS的网页导航栏的设计

针对网页导航,提出文字超链接的导航设计和利用列表实现导航的方法。给出了利用CSS实现导航条的基本方法和动态按钮导航的实现方法,同时在列表的基础上利用CSS实现了纵向导航和水平导航。通过实例,证明了基于CSS的设计导航菜单方法切实可行有效。     

基于分段的藏字校对算法研究

该文提出了一种规则和藏字语法分析相结合的藏字自动校对算法, 不使用藏字字典和大规模语料库。通过研究藏字构字语法,得到藏字的结构特征,进而对藏字的字母组合进行分段处理,简化藏字构字复杂度,研究出各分段部分的构字规则,然后按照规则进行字的校对。实验表明,系统对现代藏文字的查错率达100%。     

数值计算中的校对问题

本文说明用计算机进行数值计算时校对的必要性,就是在用高级语言编写程序时,校对仍然是必要的。同时文中还讨论了校对的一些方式和方法,并提出人工校对仍然是一种有效的方法。     

一种基于边缘的广告垃圾图像过滤方法

随着互联网的普及,电子邮件作为一种被人们广泛使用的工具,已被越来越多的不法分子用作广告、反动、色情等信息传播的工具,产生了大量的垃圾邮件.目前主流的垃圾过滤器是基于文本的,垃圾邮件制造者为了逃避被过滤,常常将文字转化为图像或者将文字嵌入到图像中,产生了大量的垃圾图像.针对广告垃圾图像多为文字图像这一特点,根据文字图像中的文字边缘分布特征,提出一种基于边缘特征的广告垃圾图像过滤方法,先检测出图像的纵向边缘,然后根据纵向边缘的分布特征提取文字行区域,最后将文字行区域去噪,确定最终的文字区域.实验证明,该方法效果良好.     

基于短语翻译模型的英语翻译计算机智能校对方法

为保证机器翻译准确性,本文提出了基于短语翻译模型的英语翻译计算机智能校对方法,通过发现可代替待校对词汇的正确词汇,以实现英语翻译的自动智能校对。同时详细设计了英语翻译计算机智能校对具体流程,分析了基于最大熵模型的分词模式。最后通过计算机智能校对代码实现与测试,得出结论,英语翻译结果校对之前准确率最高只能达到72.9%,而通过英语翻译计算机智能校对方法完成校对之后,准确率全部保持于99.0%以上,二者准确率差异显著,表明基于短语翻译模型的英语翻译计算机智能校对方法具有显著有效性与良好稳定性,值得大力推广与应用。     

Ease of annotation in proof-reading tasks

Two models of proof-reading tasks are explored by varying the procedures for annotating a text. One model assumes that the processes of detecting errors, recording annotations and resuming proof-reading are sets of serial processes. The other model assumes that annotation processes may overlap with reading the text. Performance when proof-reading a vertically displayed text (as on a CRT) and recording the errors on a separate sheet was compared with reading a horizontal text (as on a desk top) and recording annotations in the margins. The data supported the serial model and showed that variations in annotation procedures can yield differences in proof-reading speed of comparable magnitude to those found in an earlier study where CRT and printout displays were contrasted and the differences were attributed to legibility factors. The implications of these findings and this model of proof-reading are related to the wider issues of using electronic texts.     

Error categories, detection, and reduction in a musical database

The results of a detailed study of errors in a musical database are presented. It is shown that the form of analytic processing employed may substantially magnify the effect of database errors so that very small error rates can produce very large errors in the analytic results. An equation is devised to assist in estimating the effect of an overall error-rate on arbitrary analytic measures derived from the database. Considering the potential for erroneous or misleading analytic results, it is recommended that scholars using computer-based methods should habitually calculate error rates associated with their analytic procedures, and ought to present an analysis of errors in tandem with their results in order to validate their interpretations.Four methods of error detection are examined: manual proof-reading, double-entry method, programmed syntactic checking, and programmed heuristics. Significant differences were found in the thoroughness of different detection methods in uncovering all errors of a given type. The double-entry method was found to be superior to all other methods of detection; the humanities scholar's traditional allegiance to manual proof-reading was not supported by this study. Programmed methods of error detection were found to be fallible, but nonetheless useful.David Huron is a doctoral student in musicology at the University of Nottingham. He is currently engaged in research in the field of music perception.     

一个校对型汉字录入器的设计和实现

本文给出了一个校对型汉字录入器的设计和实现，详细描述了其核心－内码机制以及显示，键盘，校对和编辑模块的设计和实现。最后，作者论证了把汉字信息输入的前处理和后处理结合起来的意义。     

Proof-reading on VDUs

Two experiments are reported which compared proof-reading performance across three different modes of presentation. The results of Experiment 1 indicated that proof-reading accuracy was significantly worse on a VDU than on paper, with a photograph of the VDU display producing intermediate performance. It was also demonstrated that substitutions of visually similar errors were harder to detect than other error types. The results of Experiment 2 replicated these findings, but failed to find any difference between twin-column and single-column presentation. The findings are interpreted as indicating that character font may be a major factor in the poor performance with the VDU. It is suggested that the present experimental paradigm could be a useful assessment tool for the evaluation of display design.     

Predicting IR personalization performance using pre-retrieval query predictors

Code reviews consist in proof-reading proposed code changes in order to find their shortcomings such as bugs, insufficient test coverage or misused design patterns. Code reviews are conducted before merging submitted changes into the main development branch. The selection of suitable reviewers is crucial to obtain the high quality of reviews. In this article we present a new method of recommending reviewers for code changes. This method is based on profiles of individual programmers. For each developer we maintain his/her profile. It is the multiset of all file path segments from commits reviewed by him/her. It will get updated when he/she presents a new review. We employ a similarity function between such profiles and change proposals to be reviewed. The programmer whose profile matches the change most is recommended to become the reviewer. We performed an experimental comparison of our method against state-of-the-art techniques using four large open-source projects. We obtained improved results in terms of classification metrics (precision, recall and F-measure) and performance (we have lower time and space complexity).     

Proof-reading texts on screen and paper

This study examined the speed and accuracy of proof-reading a text presented on a CRT, relative to performance with print on paper. Two groups of 16 people each proof-read four published texts, roughly 1500 words per text For all readers, half the texts were presented as print on paper and half were presented on a 12 in. CRT screen. The two groups differed in whether the errors found in the screened text were recorded on the screen or on paper. The results suggested that the method of recording errors on the screen was quickly learned, but that both speed and accuracy were impaired when the text was presented on the screen. The implications of this for refereeing electronic journals is discussed.     

The influence of colour on CRT reading performance and subjective comfort under operational conditions

Performance and reports of comfort when proof-reading text displayed on a CRT were investigated as a function of the text/background colour. Luminances were those typically produced by commercially available hardware and software, and task conditions were designed to simulate those experienced by typical users. Colours from the extremes of the spectrum (red, blue) were shown to produce poorer reading performance, higher ratings of discomfort and a higher incidence of reported symptoms of discomfort than those from mid-spectrum and white stimulus conditions.     

Proof-reading texts on screen and paper

Abstract

This study examined the speed and accuracy of proof-reading a text presented on a CRT, relative to performance with print on paper. Two groups of 16 people each proof-read four published texts, roughly 1500 words per text For all readers, half the texts were presented as print on paper and half were presented on a 12 in. CRT screen. The two groups differed in whether the errors found in the screened text were recorded on the screen or on paper. The results suggested that the method of recording errors on the screen was quickly learned, but that both speed and accuracy were impaired when the text was presented on the screen. The implications of this for refereeing electronic journals is discussed.     

Reading music from screens vs paper

The reading of music text from a computer screen was compared to paper in a laboratory controlled study. Computer-based animated score tracking devices of three types were tested, as well as a static screen representation of the music text and its paper-based counterpart. A proof-reading exercise was given to subjects, which involved them listening to pieces of music and identifying intentional errors in the score. Their subjective views were also recorded. No significant difference between the five presentation styles were apparent in the proofreading study. However, subjects showed a significant preference for animation over paper and static representation. The most popular style of animation was where each note on the score was marked in time to the music. The medium of paper performed better overall than the static screen representation.     

Reading music from screens vs paper

The reading of music text from a computer screen was compared to paper in a laboratory controlled study. Computer-based animated score tracking devices of three types were tested, as well as a static screen representation of the music text and its paper-based counterpart. A proof-reading exercise was given to subjects, which involved them listening to pieces of music and identifying intentional errors in the score. Their subjective views were also recorded. No significant difference between the five presentation styles were apparent in the proofreading study. However, subjects showed a significant preference for animation over paper and static representation. The most popular style of animation was where each note on the score was marked in time to the music. The medium of paper performed better overall than the static screen representation.