Miyuki Kondo, Japanese Literary Research Using Corpuses: The N-gram Method and "Linguistic Resources"
Japanese Literary Research Using Corpuses: The N-gram Method and "Linguistic Resources"/p>
Miyuki Kondo
Nihongogaku, Vol.33-14
1. Introduction
In 2000, I developed a research method called "character string set operation method with N-gram" (1), and I conducted a variety of research using this method. Also, in recent years, the application of this method has proven that it is possible to perceive the phenomenon of specific language expressions being accumulated in a type of literary work, its aggregate, or language group, and to enjoy and further reproduce and spread expressions. These types of linguistic data, containing specific linguistic expressions, are called "linguistic resources" in sociolinguistics (2); however, as noted below, a variety of literary phenomena can be explained using the concept of linguistic resources. Particularly, through using a corpus and research methods such as the N-gram method, the actual condition of linguistic resources can be explored, and through this, a new and unconventional research discipline can be pioneered.
<Traditionally, in the Japanese literature and language linguistics research field, research methods employing the N-gram method had not come into common practice, and to a certain extent, there was a lack of understanding as to what kind of research it was. Accordingly, in this paper, I would first like to reintroduce corpus research with the N-gram method, and through this new concept of linguistic resources coming to light, I would like to demonstrate the completely new knowledge acquired from Japanese literature and Japanese language linguists.
