Back to results
US 8,412,712 B2Granted
Grouping methods for best-value determination from values for an attribute type of specific entity
AssigneeInternational Business Machines Corporation
InventorBarry M. Caceres
Life of this patent
Estimating the term…
Abstract
A method is provided of grouping similar values of an attribute type and determining the best value of an attribute type for an entity in an entity resolution system. By applying attribute grouping rules to organize similar values into groups, an accurate number of “distinct” values for an entity is ascertained by counting the number of resulting groups. A preferred group is selected from the resulting groups and a “best value” is selected from the preferred group by applying best value rules.
Claims14
A computer-implemented method of selecting a representative value for a specified attribute type in an entity resolution system storing identity records related to a plurality of entities, the method comprising:
- receiving a selection of an entity in the entity resolution system, wherein the selected entity is associated with a plurality of identity records stored by the entity resolution system, wherein each identity record includes one or more attribute types and corresponding attribute values, and wherein the plurality of identity records have an assumed relation to a distinct individual;
- generating one or more distinct value groups, wherein each distinct value group includes one or more attribute values corresponding to the specified attribute type, wherein generating the one or more distinct value groups comprises:
- identifying a set of attribute values corresponding to the specified attribute type;
- calculating a pair-wise comparison score for each distinct attribute value pair in the set of attribute values;
- and
- assigning each distinct attribute value pair in the set of attribute values to one of the one or more distinct value groups based upon the pair-wise comparison score calculated for each distinct attribute value pair in the set of attribute values, wherein a respective distinct attribute value pair is assigned to the one of the one or more distinct value groups upon determining that its pair-wise comparison score is within a specified grouping threshold;
- selecting one of the one or more distinct value groups as a preferred group;
- and
- selecting, from the preferred group, a first attribute value as the representative value for the specified attribute type of the selected entity by selecting an attribute value with a least average pair-wise difference with other attribute values of the preferred group, wherein selecting the attribute value with the least average pair-wise difference comprises analyzing the pair-wise comparison score calculated for each distinct attribute value pair in the preferred group.
6 more elements
A computer program product for selecting a representative value for a specified attribute type in an entity resolution system storing identity records related to a plurality of entities, the computer program product comprising a non-transitory computer usable storage medium having computer usable program code configured to:
- receive a selection of an entity in the entity resolution system, wherein the selected entity is associated with a plurality of identity records stored by the entity resolution system, wherein each identity record includes one or more attribute types and corresponding attribute values, and wherein the plurality of identity records have an assumed relation to a distinct individual;
- generate one or more distinct value groups, wherein each distinct value group includes one or more attribute values corresponding to the specified attribute type, wherein generating the one or more distinct value groups comprises:
- identifying a set of attribute values corresponding to the specified attribute type;
- calculating a pair-wise comparison score for each distinct attribute value pair in the set of attribute values;
- and
- assigning each distinct attribute value pair in the set of attribute values to one of the one or more distinct value groups based upon the pair-wise comparison score calculated for each distinct attribute value pair in the set of attribute values, wherein a respective distinct attribute value pair is assigned to the one of the one or more distinct value groups upon determining that its pair-wise comparison score is within a specified grouping threshold;
- select one of the one or more distinct value groups as a preferred group;
- and
- select, from the preferred group, a first attribute value as the representative value for the specified attribute type of the selected entity by selecting an attribute value with a least average pair-wise difference with other attribute values of the preferred group, wherein selecting the attribute value with the least average pair-wise difference comprises analyzing the pair-wise comparison score calculated for each distinct attribute value pair in the preferred group.
6 more elements
A system comprising:
- a processor;
- and
- a memory containing a program, which, when executed by the processor, is configured to select a representative value for a specified attribute type in an entity resolution system storing identity records related to a plurality of entities by performing an operation comprising:
- receiving a selection of an entity in the entity resolution system, wherein the selected entity is associated with a plurality of identity records stored by the entity resolution system, wherein each identity record includes one or more attribute types and corresponding attribute values, and wherein the plurality of identity records have an assumed relation to a distinct individual;
- generating one or more distinct value groups, wherein each distinct value group includes one or more attribute values corresponding to the specified attribute type, wherein generating the one or more distinct value groups comprises:
- identifying a set of attribute values corresponding to the specified attribute type;
- calculating a pair-wise comparison score for each distinct attribute value pair in the set of attribute values;
- and
- assigning each distinct attribute value pair in the set of attribute values to one of the one or more distinct value groups based upon the pair-wise comparison score calculated for each distinct attribute value pair in the set of attribute values, wherein a respective distinct attribute value pair is assigned to the one of the one or more distinct value groups upon determining that its pair-wise comparison score is within a specified grouping threshold;
- selecting one of the one or more distinct value groups as a preferred group;
- and
- selecting, from the preferred group, a first attribute value as the representative value for the specified attribute type of the selected entity by selecting an attribute value with a least average pair-wise difference with other attribute values of the preferred group, wherein selecting the attribute value with the least average pair-wise difference comprises analyzing the pair-wise comparison score calculated for each distinct attribute value pair in the preferred group.
9 more elements
Dependencies are inferred from the claim text.