| 摘要 |
PURPOSE:To preferentially convert words belonging to the same field as that of the input numerical expression or chemical formula by discriminating the field of the numerical expression or the chemical formula and converting preferentially the character strings of the decided field. CONSTITUTION:A microprocessor 1 applies the morpheme analysis, the syntactic analysis, etc., to the phonetic reading inputted from a keyboard 4 and divides the reading into paragraphs to decide the input or non-input of a numerical expression or a chemical formula. If the numerical expression or the chemical formula is inputted, the KANA (Japanese syllabary)/KANJI (Chinese character) conversion is carried out. Then the most likely word is decided as the first candidate and its homonym is produced. When the chemical formula is inputted, the most likely word is decided as the first candidate among those words of the chemical field attribute based on the field information stored in the dictionary contained in a ROM 2. When the numerical expression is inputted, the most likely word is decided as the first candidate among those words of the numerical field attrinute. |
