| 摘要 |
<P>PROBLEM TO BE SOLVED: To provide a method and system for identifying and analyzing commonly confused word with a natural language parser. <P>SOLUTION: A computer system parses an input text consisting of more than one word by using a relation of mapping from potentially confused words including one word out of the words of the input text into words considered to be intended. The computer system in the first place identifies possible word class to each word of the input text including potentially confused words, next identifies possible word classes to the words considered to be intended which the above relation maps from the potentially confused words, and finally generates a complete syntactic tree including the word classes to the intended words but not a complete syntactic tree including the word classes to the potentially confused words by applying a syntactic grammar rule to those identified word classes. <P>COPYRIGHT: (C)2007,JPO&INPIT |
