| 摘要 |
PROBLEM TO BE SOLVED: To enable efficient Japanese syllabary (KANA)/Chinese character (KANJI) conversion to a long sentence by providing a learning buffer for continuously storing plural KANA and correspondent KANA/KANJI CONVERTED RESULTS. SOLUTION: A 1st converting processing part 12 converts a KANA string to KANJI. A KANJI buffer 13 stores the KANA/KANJI converted result. A long sentence KANA buffer 14 can store the contents of plural character strings in the KANA buffer 11 inputted to the 1st converting processing part 12. A long sentence KANJI buffer 15 continuously stores the contents of plural outputs in a KANJI buffer 13 at the 1st converting processing part 12 just for character strings corresponding to the number of character strings in the long sentence KANA buffer 14. The learning buffer is composed of a KANA learning buffer 16 and a KANJI learning buffer 17 for coupling the character strings in the long sentence KANA buffer 14 and the long sentence KANJI buffer 15 as one pair and storing the contents of these character strings. A 2nd converting processing part 18 refers to the character strings in the KANA learning buffer 16, retrieves the KANJI learning buffer 17 corresponding to the character string in the KANA learning buffer 16 and outputs the result as a converted character string. |
