Introduction

Method to create Long Unit Word from an automatic division of a Japanese text into morphemes. Each LUW is associated with a correct transcription in the phonetic alphabet hiragana and transliterated into Latin characters, facilitating the learning of the written language for both Japanese children and foreigners of all ages, allowing easier identification of word boundaries and the associated grammar rules.

Technical features

Japanese text does not have spaces between synograms and uses 3 main types of characters: 2 phonetic alphabets: – hiragana and katakana and kanji synograms. At the state of the art, there are systems that allow to translate a Japanese text into morphemes or that create LUW from dictionaries of morphemes, but it was not yet possible to properly automate a process that, even when done manually, is a source of errors and misinterpretations. The proposed method allows the automatic creation of LUW, their correct transcription in kana and a transliteration in Latin characters. This offers the possibility to present the same Japanese text in multiple versions, allowing the understanding of the same by learners with different levels of knowledge of the language. The method has been deposited as Computer Implemented Invention and is the basis of an online platform for the study of the Japanese language.

Possible Applications

• Japanese Language E-Learning method;
• Automatic translation systems;
• Automatic text reading systems.

Advantages

• Facilitates understanding of Japanese text;
• Automates the process of creating LUW;
• Automates transcription between phonetic alphabets and transliteration into Latin characters.