日本江戶初期漢字異體字知識庫的建立

使用漢字是漢字文化圈的共同特色，其中，日本不僅使用漢字已經超過一千五百年，對於漢字的研究和整理也有許多成果。在目前文獻中，大部份的中日漢字資料庫，多數只能呈現統一碼（Unicode）所收的字形，但是缺乏運用中日雙方的字書內容，將其表達在計算機資料庫中，以致於中日漢字資料庫沒有更深層的漢字知識，難以作為研究中日漢字和異體字的知識庫。本研究的目是設計與建立一個日本異體字知識系統，能夠表達日本異體字字書的知識，期望成為協助中日漢字異體字研究的資源。江戶時期為日本研究異體字最興盛的時期，本研究以日本教育漢字及其異體字為對象，完成《異字篇》、《異體字辨》、《正俗字例》、《倭楷正訛》和《道齋隨筆》等江戶時期異體研究的知識庫，並且與中文漢字知識本體的整合，用字覆蓋率達89.14%，不僅可提供中日漢字異體研究，也改善了過去中日漢字資料庫只有編碼系統字形的侷限。

關鍵字

日本漢字；日本異體字；異體字資料庫；中日漢字比較

並列摘要

Using Chinese characters is a common feature in Sinosphere. For instance, Japan has been using Chinese characters (Kanji) for the past 1500 years and has accomplished a variety of researches on Chinese characters. In previous relative studies, only Unicode fonts which are present in the Hanzi databases have been applied, but this database was in lack of knowledge on researches of Chinese and Japanese variants. Consequently the comparison of Chinese and Japanese variants has not been integrated into the knowledge base. As a result, it is difficult to be a knowledge base for studying Chinese and Japanese variants. The objective of this paper is to propose a knowledge system which can represent the results of studying the variants of the early Edo period for the comparison of Chinese and Japanese variants. This research uses Japan educational characters and their variants as objects, which represented the knowledge from Iji hen, Itai jiben, Seizoku jirei, Wakai saika and Do sai zuihitsu. The results of this study cover 89.14% usage and have also been integrated with Hantology. The main contributions of this paper not only provide a knowledge base for the comparison of Chinese and Japanese variants but also break through the limitation of previous studies on Chinese and Japanese databases.