| 摘要 |
<p>The objective of the present invention is to accurately assess association among data on the basis of incomplete conceptual information in which not all information has been registered. This system for assessing association among data is provided with: candidate selection means which, on the basis of one or a plurality of concept collections which have been generated, if a character string of the same concept is included among data that are compared, or if one of the data units includes a character string belonging to the concept, selects the samedata as a candidate for similarity calculation; and similarity calculation means which performs calculation processing of similarity for the candidate which has been selected at the candidate selection means and, meanwhile, sets the similarity for the candidate that was not selected at the candidate selection means to a pre-defined small value so as to output the similarity between the data which are compared.</p> |
