| 摘要 |
PROBLEM TO BE SOLVED: To generate a classification parameter whose classification accuracy is improved. SOLUTION: The data classification processing apparatus 100 includes: an input processing part 101 to input learning data which is previously classified into each category; a feature generating part 105 for generating feature for each classification category from the learning data; a feature weight calculating part 111 for calculating feature weight for each feature of each classification category; a classification parameter generating part 107 for generating a classification parameter to be a reference for classifying data to be later input for each classification category on the basis of weight of each feature; a classification parameter determining part 108 for determining whether the generated classification parameter can classify the learning data for each classification category; and a classification parameter combining part 109 for combining the classification parameter of a different classification category with another classification parameter when it is determined that the learning data is classified into a classification category different from a previously classified classification category. COPYRIGHT: (C)2009,JPO&INPIT
|
