基于NLB检索系统词语搭配统计值的日语近义词词典编纂构想

Conceptions of Japanese Synonym Lexicography Based on the Statistics of Word Collocation in NLB Retrieval System

日语近义词词典作为日语学习者重要的参考工具书,若不能提供丰富、详尽、科学的近义词辨析信息,很可能导致误用,阻碍学习。语料库提供大量的真实语料,基于语料库的词语搭配,对近义词进行辨析,使近义词研究更具科学性和有效性。NLB检索系统是日本国立国语研究所和Lago语言研究所共同开发的日语在线检索系统,检索数据源于「現代日本語書き言葉均衡コーパス」语料库。NLB自动呈现节点词与搭配词的共现频数、MI值和LD值。本文基于NLB检索系统,以近义词「文句」和「苦情」为例,从词配角度区分二者的异同,探讨日语近义词词典编纂的新方法。

As an important reference book for Japanese learners, Japanese synonym dictionary will impede students＇ Japanese learning and lead them to misuse some words if it cannot provide rich, detailed and scientific synonym discrimination information. Since a corpus provides a large quantity of authentic language materials, synonym discrimination based on a corpus can make the research into synonyms more scientific and effective. As an online Japanese retrieval system jointly developed by National Institute for Japanese Language and Linguistics and Lago Language Research Institute, NLB gets its data from the Balanced Corpus of Contemporary Written Japanese and it can automatically reflect the co-occurrence frequency, MI value and LD value of node words and their collocates. Based on the collocation information found in NLB, this paper analyzes the similarities and differences between ＂ 文句＂ and ＂苦情＂, and thus explores the new methods of Japanese synonym lexicography.