| 摘要 |
PROBLEM TO BE SOLVED: To achieve high classification accuracy by the small number of classification parameters as much as possible in a data classification method with a teacher based on Multinomial Bayes. SOLUTION: A feature generation part 11 generates features from a learning text input through an input/output device 2. A feature selection part 12 inputs a feature group generated by the feature generation part 11 and selects features supposed as effective for classification from the feature group to generate a classification parameter candidate group. A classification parameter selection part 131 in a machine learning part 13, while giving weight to the classification parameter candidate group on the basis of machine learning algorithm based on Multinomial Bayes, determines whether the classification parameter candidate group is effective as classification parameters from the given weight, and outputs only effective classification candidate parameters as classification parameters. COPYRIGHT: (C)2011,JPO&INPIT
|
