* Methods of Morphological Analysis of English
Unlike Japanese, English sentences are clearly delimited in most places beforehand. Therefore, the word segmentation process does not need to be as complex as in Japanese, and is often based on simple rules.
Morphological analysis, which is used as a natural language processing technique, is a method of determining the morphemes that make up a sentence by comparing sentences containing complex Japanese grammar and words with the dictionary information registered in morphological analysis tools.
For an AI to respond to a human question, it must first understand the meaning of the words in question. Natural language processing is the technology required for this, and "morphological analysis" is the most fundamental part of this natural language processing technology.
Furthermore, Japanese is said to be less logical than English. This is due, first of all, to the sentence structure in which the subject is followed by the predicate. Secondly, the sentence structure has a conclusion at the end.
Although the Japanese language is considered unsuitable for such text mining, many Japanese language analysts have made various attempts.
This time, I tried to analyze the Japanese text using a python program that had been made available to the public. However, it is not meant for academic purposes.
The object of analysis was a modern translation of the Japanese classic "The Tale of Genji".
Create an analysis program for 5cho wakamurasaki by referring to the co-occurrence net program in Iruka no Box:いるかのボックス: Pythonで共起ネットワークを作成する (irukanobox.blogspot.com)
Next, this was extended to the other books and simply superimposed so that it could be developed three-dimensionally. Therefore, there is no correlation between the previous and next books, nor is there a common factor linkage between each book.
