| 摘要 |
<p><P>PROBLEM TO BE SOLVED: To obtain a similarity analysis evaluation system which extracts a feature amount obtained by paying attention to the shape of time-series data about a large amount of time-series data, classifies the large amount of time-series data arbitrarily on the basis of the extracted feature amount and achieves goodness-of-fit evaluation based on attributes and classification results of the time-series data. <P>SOLUTION: A feature amount extracting means 6 extracts a feature amount obtained by paying attention to the shape of corresponding time-series data by using the time-series data 5 extracted from time-series data 3 having attributes of a consumer by a retrieval-tabulation engine 4, a clustering means 7 classifies the feature amount extracted by the feature amount extracting means 6 into a plurality of data sets having a similarity feature, and in addition, an evaluation matrix generating means 8 generates an evaluation matrix for evaluating results classified by the clustering means 7 on the basis of the attributes of the consumer belonging to the time-series data. <P>COPYRIGHT: (C)2010,JPO&INPIT</p> |
