| 主权项 |
1. A computer-implemented method for comparing similarity between an entity of interest and a candidate entity, said method comprising:
a) retrieving from a data repository data records representing said entity of interest and said candidate entity, said data records include at least attribute data, scale data and group operation data, wherein the attribute data representing attributes of said entity of interest and said candidate entity, the scale data representing scales associated with respective attributes and signifying levels of significance of said respective attributes, said attributes being classified into groups, said group operation data being indicative of different group operations within said groups, said attributes associated with said scales forming predetermined shapes on a Graphic User Interface (GUI) representing said entity of interest and said candidate entity; b) comparing scale data associated with each attribute of said entity of interest with scale data associated with the same attribute of said candidate entity, selecting the smaller scale of the comparison as Attribute Compare Score for each attribute; c) calculating a Group Score for each group of said groups, including: if the group operation data of a group is set to OR, calculating a ratio of the largest Attribute Compare Score within said group to the maximum scale of attributes of the group, setting the ratio as the Group Score; otherwise the group operation data of the group is set to AND, calculating a ratio of a sum of Attribute Compare Scores of the group to a sum of the scale data associated with the attributes of said entity of interest, setting the ratio as the Group Score; and d) summing up Group Scores of said groups as a Shape Match Score, representing a similarity indication between said entity of interest and said candidate entity. |
