| 摘要 |
PROBLEM TO BE SOLVED: To eliminate input errors and to eliminate erroneous descriptions in the case of preparing a handwritten document by reporting a correct reading to an operator in the case of allowing the erroneous input of a reading and performing conversion at the time of KANA (Japanese syllabary)/KANJI (Chinese character) conversion or the like. SOLUTION: A document data storage area 2-1, a non-converted character storage buffer 2-2 and a Japanese conversion work area 2-3 are allocated to a RAM as memory areas, a CPU syntax-analyzes an inputted reading character string corresponding to a program for controlling the Japanese conversion processing of continuous clauses and converts the reading character string to KANJI candidates or KANJI/KANA mixed candidates for each character node and a Japanese conversion dictionary 3-2 is referred to. The Japanese conversion dictionary 3-2 stores a conversion clause candidate information group, each conversion clause candidate information is composed of the reading, notation and part-of-speech information and an 'erroneous reading' is stored other than the 'correct reading' for the reading to be easily mistaken. Then, the conversion for allowing the erroneous input of the reading is performed and the CPU window-displays the correct and erroneous readings and clearly indicates them to the operator. |
