| 摘要 |
<p><P>PROBLEM TO BE SOLVED: To provide a device, a method, and a program for data analysis, capable of deriving a causal relationship of an output attribute and an input attribute without defining a label hierarchical structure in advance. <P>SOLUTION: This data analysis device has a data classification part 4 classifying a master data group into a first data group and a second data group according to an output attribute, a frequency cumulative difference calculation part 7 calculating a frequency cumulative difference indicating the degree that the data whose input attribute is below the value inclines toward either the first data group or the second data group for all values of each input attribute, an input attribute threshold deciding part 8 deciding a value with the greatest frequency cumulative difference about each input attribute as a threshold of each input attribute, and a frequency cumulative ratio computing part 16 calculating a 2-J ratio for each input attribute. It is also provided with a factor extraction part 9 extracting an input attribute condition with the highest 2-J ratio as information indicating a factor of an output attribute condition corresponding to the second data group. <P>COPYRIGHT: (C)2005,JPO&NCIPI</p> |
