Entity expansion and grouping

1. A non-transitory computer readable storage medium, comprising executable instructions to:

receive a set of entities;

process each of the entities in a single pass of the set of entities, the processing for each entity comprising:

converting the entity to a standard form including normalized attributes, a tag reference and a feature;

expanding the entity with corresponding variants;

forming an annotated entity combining the standard form and the corresponding variants; and

assigning, using an entity grouper, the entity to (i) an existing group generated when processing a previous entity that matches the annotated entity and (ii) a new group if the annotated entity does not match an existing group;

wherein:

the entity grouper is implemented with a set of state machines and, at least a portion of the entities are converted into regular expressions that are compiled into finite state networks processed by finite state transducers,

each annotated entity comprises:

a feature of the corresponding entity in an originally observed form in a corresponding data set,

an annotation of the entity in a normalized form having tag delimited components, and

at least one variant of the normalized form having tag delimited components.

2. The computer readable storage medium of claim 1 wherein the normalized attributes include normalized punctuation, normalized case and normalized abbreviations.

3. The computer readable storage medium of claim 1 wherein the tag reference characterizes the entity.

4. The computer readable storage medium of claim 1 wherein the feature characterizes an attribute of the entity.

5. The computer readable storage medium of claim 1 wherein the corresponding variants characterize presumed equivalent entities.

6. The computer readable storage medium of claim 1 wherein the group specifies the standard form for a first instance.

7. The computer readable storage medium of claim 6 wherein the group specifies a second instance.

8. The computer readable storage medium of claim 7 wherein the group specifies a third instance with a conjectured inclusion in the group.

9. The computer readable storage medium of claim 1 wherein matching used in connection with the assigning is based upon a standard form match.

10. The computer readable storage medium of claim 1 wherein matching used in connection with the assigning is based upon a variant form match.

11. The computer readable storage medium of claim 1 wherein matching used in connection with the assigning is based upon a conjectured match, the conjectured match being based on a proximity between the annotated entity and the matching group when a full standard form match or a variant form match does not exist.

12. The computer readable storage medium of claim 1 further comprising executable instructions to form a plurality of groups.

13. The computer readable storage medium of claim 12 further comprising executable instructions to analyze relationships between the plurality of groups.

14. A method for implementation by one or more data processors comprising:

receiving, by at least one data processor, a set of entities;

processing, by at least one data processor, each of the entities in a single pass of the set of entities, the processing for each entity comprising:

converting the entity to a standard form including normalized attributes, a tag reference and a feature;

expanding the entity with corresponding variants;

forming an annotated entity combining the standard form and the corresponding variants; and

assigning, using an entity grouper, the entity to (i) an existing group generated when processing a previous entity that matches the annotated entity and (ii) a new group if the annotated entity does not match an existing group;

wherein:

the entity grouper is implemented with a set of state machines and, at least a portion of the entities are converted into regular expressions that are compiled into finite state networks processed by finite state transducers,

each annotated entity comprises:

a feature of the corresponding entity in an originally observed form in a corresponding data set,

an annotation of the entity in a normalized form having tag delimited components, and

at least one variant of the normalized form having tag delimited components.

15. A system comprising:

one or more data processors; and

memory storing instructions which, when executed, result in operations comprising:

receiving, by at least one data processor, a set of entities from at least one data set;

processing, by at least one data processor, each of the entities in a single pass of the set of entities, the processing for each entity comprising:

converting the entity to a standard form including normalized attributes, a tag reference and a feature;

expanding the entity with corresponding variants;

forming an annotated entity combining the standard form and the corresponding variants; and

assigning, using an entity grouper, the entity to (i) an existing group generated when processing a previous entity that matches the annotated entity and (ii) a new group if the annotated entity does not match an existing group;

wherein:

the entity grouper is implemented with a set of state machines and, at least a portion of the entities are converted into regular expressions that are compiled into finite state networks processed by finite state transducers,

each annotated entity comprises:

a feature of the corresponding entity in an originally observed form in the corresponding data set,

an annotation of the entity in a normalized form having tag delimited components, and

at least one variant of the normalized form having tag delimited components.

16. The system of claim 15 wherein matching used in connection with the assigning is based upon a variant form match.

17. The system of claim 15 wherein matching used in connection with the assigning is based upon a conjectured match, the conjectured match being based on a proximity between the annotated entity and the matching group when a full standard form match or a variant form match does not exist.

18. The method of claim 14 wherein matching used in connection with the assigning is based upon a standard form match.

19. The method of claim 14 wherein matching used in connection with the assigning is based upon a variant form match.

20. The method of claim 14 wherein matching used in connection with the assigning is based upon a conjectured match, the conjectured match being based on a proximity between the annotated entity and the matching group when a full standard form match or a variant form match does not exist.