| 摘要 |
PROBLEM TO BE SOLVED: To efficiently construct an ontology optimum to each user.SOLUTION: A device for processing a command according to the present invention that processes a voice command comprises: similarity storage means for storing a similarity that is a degree of similarity between independent words whose meanings are similar; conversion means for converting a voice input by a user into an input command that is character data; command candidate generation means for generating, in reference to the similarity storage means, a candidate for a recognition command capable of being recognized by the device for processing a command on the basis of independent words contained the input command; and update means for updating the similarity storage means so that similarities can be increased between independent words contained in the recognition command and independent words contained in the input command, when the generated candidate for the recognition command is uniquely determined. |
