Study to build full text data base of Medieval Japanese Documents.

点击翻译按钮获取中文摘要

Recently we can use the computer technology to treat historical documents as digital data. But it seems necessary to develop and test the measures to process these documents in publication by computer. Because they have too various and complex forms to be input on the unique record format.At first we use the HTML as input format and the fulltext retrieval system as main database. As a result of this, we are able1) to use the published content list in order to make the hyper links to the relevant text.2) to build KWIC retrieval system on the Web site of Historiographical Institute University of Tokyo.3) to accumulate historical terms in Japanese medieval era as a corpus.4) to make a list of the special characters ( "gaiji" ) necessary for managing medieval Japanese documents (so-called "Hissu-kanji-hyo" ).The result is now open on the homepage of the Historiographical Institute. Onto this site, we load up two databases. That is,1) Full text Databese of Japanese manuscripts ( "komonjo" ). Now consist of all content list of "Dainihon-Komonjo-Iewake" ( "Japanese Historical manuscripts, each possessors", compiled and published by Historiographical Institute), and the text of several books extracted from same series.2) Full text Database of historical materials in Heian-era. Now consist of "Heian-Ibun" ( "All Heian documents", compiled by Rizo Takeuchi and published by Tokyo-do), and the text of several books extracted from "Dainihon-Kokiroku" ( "Japanese Historical Diaries", compiled by Historiographical Institute and published by Iwanami-shoten).