Unguided curiosity in support of entity resolution techniques

1. A method, comprising:

receiving, using a processor of a computer, data comprising an entity having a feature;

determining how the entity correlates with an existing entity;

generating an entity resolution assertion having a confidence score;

determining that there is a consequence based on the confidence score and the entity resolution assertion being wrong;

identifying an additional feature for the entity to increase the confidence score of the entity resolution assertion, wherein the additional feature is not received with the entity;

searching an external data source for the additional feature to obtain an observation containing the additional feature; and

performing entity resolution using the at least one feature and the additional feature.

2. The method of claim 1 , further comprising:

evaluating how the received entity correlates with the existing entity to determine whether a consequence threshold is met; and

in response to the consequence threshold being met, identifying the additional feature.

3. The method of claim 1 , further comprising:

selecting one or more data sources in which to search for the additional feature;

prioritizing the one or more data sources; and

selecting the external data source from the prioritized one or more data sources.

4. The method of claim 1 , wherein the additional feature enables two records to be resolved to one entity or multiple records in one entity to be split.

5. The method of claim 1 , further comprising:

discovering a new data source in which to search for the additional feature, wherein the external data source is the discovered, new data source; and

recording the new data source that has been discovered.

6. The method of claim 1 , further comprising:

based on a kind of entity being compared, identifying one or more additional features known to describe the kind of entity; and

outputting the one or more additional features.

7. The method of claim 1 , wherein identifying the additional feature further comprises:

identifying multiple additional features;

prioritizing the multiple additional features to be searched based on cost and accuracy; and

selecting the additional feature from the prioritized additional features.

8. The method of claim 1 , wherein the data source is one of prescribed and self discovered.

9. A computer system, comprising:

a processor; and

a storage device coupled to the processor, wherein the storage device stores a program, and wherein the processor is configured to execute the program to perform operations, the operations comprising:

receiving data comprising an entity having a feature;

determining how the entity correlates with an existing entity;

generating an entity resolution assertion having a confidence score;

determining that there is a consequence based on the confidence score and the entity resolution assertion being wrong; and

identifying an additional feature for the entity to increase the confidence score of the entity resolution assertion, wherein the additional feature is not received with the entity;

searching an external data source for the additional feature to obtain an observation containing the additional feature; and

performing entity resolution using the at least one feature and the additional feature.

10. The computer system of claim 9 , wherein the operations further comprise:

evaluating how the received entity correlates with the existing entity to determine whether a consequence threshold is met; and

in response to the consequence threshold being met, identifying the additional feature.

11. The computer system of claim 9 , wherein the operations further comprise:

selecting one or more data sources in which to search for the additional feature;

prioritizing the one or more data sources; and

selecting the external data source from the prioritized one or more data sources.

12. The computer system of claim 9 , wherein the additional feature enables two records to be resolved to one entity or multiple records in one entity to be split.

13. The computer system of claim 9 , wherein the operations further comprise:

discovering a new data source in which to search for the additional feature, wherein the external data source is the discovered, new data source; and

recording the one new data source that has been discovered.

14. The computer system of claim 9 , wherein the operations further comprise:

based on a kind of entity being compared, identifying one or more additional features known to describe the kind of entity; and

outputting the one or more additional features.

15. The computer system of claim 9 , wherein the operations for identifying the additional feature further comprise:

identifying multiple additional features;

prioritizing the multiple additional features to be searched based on cost and accuracy; and

selecting the additional feature from the prioritized additional features.

16. The computer system of claim 15 , wherein the data source is one of prescribed and self discovered.

17. A computer program product comprising a computer readable storage device including a computer readable program, wherein the computer readable program when executed by a processor on a computer cause the computer to:

receive data comprising an entity having a feature;

determine how the entity correlates with an existing entity;

generate an entity resolution assertion having a confidence score;

determine that there is a consequence based on the confidence score and the entity resolution assertion being wrong;

identify an additional feature for the entity to increase the confidence score of an entity resolution, wherein the additional feature is not received with the entity;

search an external data source for the additional feature to obtain an observation containing the additional feature; and

perform entity resolution using the at least one feature and the additional feature.

18. The computer program product of claim 17 , wherein the computer readable program when executed by the processor on the computer cause the computer to:

evaluate how the received entity correlates with the existing entity to determine whether a consequence threshold is met; and

in response to the consequence threshold being met, identifying the additional feature.

19. The computer program product of claim 17 , wherein the computer readable program when executed by the processor on the computer cause the computer to:

select one or more data sources in which to search for the additional features;

prioritize the one or more data sources; and

select the external data source from the prioritized one or more data sources.

20. The computer program product of claim 17 , wherein the additional feature enables two records to be resolved to one entity or multiple records in one entity to be split.

21. The computer program product of claim 17 , wherein the computer readable program when executed by the processor on the computer cause the computer to:

discover a new data source in which to search for the additional feature, wherein the external data source is the discovered, new data source; and

record the new data source that has been discovered.

22. The computer program product of claim 17 , wherein the computer readable program when executed by the processor on the computer cause the computer to:

based on a kind of entity being compared, identifying one or more additional features known to describe the kind of entity; and

outputting the one or more additional features.

23. The computer program product of claim 17 , wherein, when identifying the additional feature, the computer readable program when executed by the processor on the computer cause the computer to:

identify multiple additional features;

prioritize the multiple additional features to be searched based on cost and accuracy; and

select the additional feature from the prioritized additional features.

24. The computer program product of claim 23 , wherein the data source is one of prescribed and self discovered.