英文摘要 |
A central area of the ongoing third AI boom is natural language processing technology for processing human language. For humanities research, which has not been closely associated with information technology, the development of natural language processing is not only a threat to the replacement of language-related work, but also an opportunity to develop into a new field. This paper discusses the rapidly growing text mining technology in the field of natural language processing and considers the procedures and cautions for its application in humanities research on Japanese language. There are many essential differences in research directions and methods between the humanities, the natural sciences, and the social sciences, and imitating the methods of the natural and social sciences in the humanities will not produce any results. When text mining technology is applied to the humanities, the problem of unification of quantitative and qualitative methods is inevitable, and researchers who have not been exposed to information processing technology have a major problem in the field of pre-processing, which they can easily stumble upon. As a procedure, this paper describes the procedures and cautions related to pre-processing, then considers how to conduct analysis and apply the results of text mining to the study of the humanities, and presents a new perspective on the application of text mining technology to the humanities, using Haruki Murakami's Pinball, 1973 as a concrete example of analysis. We have shown that these processes are possible. We hope that by pioneering application methods, we can pave the way for the application of AI technology to research and education in the field of Japanese language and humanities in Taiwan. |