Preliminary Work for Lexicographical Study of Ainu Epic

本计画的主要目的是为未来的阿伊努史诗辞典编纂做准备，第一年（1992年）将金大一京介与久保寺逸彦所编纂的阿伊努史诗典籍全部整理成资料库（120，000行）。第二年（1993年），在UNIX上编写了一个程序，它从输入的文本中提取每个单词及其上下文，并自动按字母顺序排列。我们对每一首史诗都有索引。今年，我们选择了其中的一个索引作为样本，开始为最著名的7000行史诗《板取丸》编纂词典。调查员分析了每一个词，得出了它的成分。第三年（1994年），调查员的工作集中在两个主题上。第一步是分析每个词的所有语境，并根据语境重新确定词的位置。这项工作仍在进行中。第二个是程序的修订，因为旧版本运行太慢，无法分析12万行史诗（约4,000天）。这个问题在明年得到解决（大约10小时）。

It is the purpose of this prelinary project to compile Ainu Epic Lexicon in the future.In the first year (1992) all the classical texts of the Ainu epic edited by Kyoosuke KINDAICHI and Itsuhiko KUBODERA were databanked (120,000 lines).In the second year (1993) a program on UNIX was made, which picked up every word with its context from the inputted texts and arranged them in alphabetical order automatically. We got a concordance for every piece of epics with this program. In this year we chose one of these concordances as a sample and began to compile a lexicon to "Itadori-Maru", the most famous epic with 7000 lines.Ainu is a incorporative and polysynthetic language. The investigator analyzed every word and got its constituents. These constituets should also be listed in the lexicon.In the third year (1994) the investigator's work was concentrated on two subjects. The first was analyzeing all the contexts of every word and rearrangeing occurences of the words contextually. This work is still under way. The second was revisal of the program, for the old version ran too slow to analyze the 120,000 lines of epics (about 4,000 days). This problem was solved next year (about 10 hours).