Making of CD-ROM of Research Papers of Quantitative Japanese Linguistics and Development of the Technology of its Usages

1.我们准备了一个初步试用版的光盘日本定量论文。然后，我们招募了一些监督员，请他们评估光盘并给我们一些建议。我们考虑了光盘的一些有效和有用的用法。我们使用了五个光学字符阅读器软件，并将一些扫描的纸张放入软件中，以评估OCR软件是否可以实际用于字符识别过程。结果表明，即使是最有效的软件也不能完全识别我们的文档。误差水平是一些不确定因素的顺序。事实上，OCR软件远没有达到实用的成功水平，我们不能把这些软件作为一种输入法。我们调查了每个研究领域的数百篇研究论文，并试图了解每篇论文的数据量。我们将硬拷贝的研究论文按照其领域（词汇、社会语言学、字母、方言、语音和语法）分类，并检查每个数据的数量。结果表明，各篇论文的数据量存在很大的差异，很难把握数据量的变化趋势。这是因为每一项研究都有其不同于其他论文的特定目的。

1. We prepared a preliminary trial version of CD-ROM of Japanese Quantitative papers. We then recruited some monitors and we asked them to evaluate the CD-ROM and to give us some suggestions. We considered some valid and useful usage of the CD-ROM.2. We utilized five Optical Character Reader softwares and put some scanned papers into the softwares in order to evaluate OCR softwares whether they can be practically used in the process of character recognition or not. The result showed that even the most effective software cannot recognize our documents completely. The error level is some percents order. The fact says that OCR softwares are far from practically successful level, and we cannot utilize those softwares as an input method.3. We surveyed hundreds of research papers in each research field, and tried to realize the quantity of data of each paper. We classified hard-copied research papers which will be stored in the CD-ROM according to its field (vocabulary, sociolinguistics, letters, dialect, phonetics, and grammar), and checked the quantity of each data. The result indicated that the quantity of data of each paper is full of variance, and we found it difficult to grasp a tendency of quantity. The reason is that each research has its own specific purpose different from that of other papers.