| 摘要 |
<p>A representative feature extraction system for selecting a representative feature from an input data group is provided with an occurrence distribution memorizing means for memorizing occurrence distributions regarding feature amounts which are assumed to be input, an evaluation value calculation means for obtaining an evaluation value of each piece of data in the data group by calculating a sum of distances to other pieces of data in the data group on the basis of the occurrence distributions, and a data selection means for selecting a piece of data having the minimum evaluation value as a representative feature of the data group.</p> |
