Unstructured data parsing for structured information

1. A method comprising:

receiving unstructured text;

identifying a plurality of text segments within the unstructured text and enriching the unstructured text by assigning one or more metadata labels to each of the text segments;

calculating one or more feature values for each text segment for at least one of the metadata labels associated with at least one of the text segments;

providing the feature values, metadata labels, and text segments to a trained machine learning system;

receiving, from the machine learning system, an indication of a classification for at least one of the text segments;

based upon the indication of the classification for the at least one of the text segments, identifying the at least one of the text segments as a portion of contact data;

receiving, from the machine learning system, a confidence score for each of the classifications, wherein the at least one of the text segments is identified as a portion of contact data based upon a confidence score assigned to the classification of the at least one text segment; and

generating structured contact information for an entity based on the confidence score for each of the classifications and the text segments.

2. The method of claim 1 , wherein the step of assigning one or more metadata labels to each of the text segments comprises:

processing each text segment of the text segments with at least one enrichment source; and

based upon an output of the at least one enrichment source, assigning the one or more metadata labels to the each text segment.

3. The method of claim 2 , wherein the at least one enrichment source is selected from the group consisting of: a dictionary of name parts; a pattern matching system; an address-matching library; and an external contact information service.

4. The method of claim 3 , wherein the at least one enrichment source comprises a plurality of different enrichment sources, each enrichment source independently selected from the group consisting of: a dictionary of name parts; a pattern matching system; an address-matching library; and an external contact information service.

5. The method of claim 1 , wherein the structured contact information includes only text segments having a confidence score above a minimum confidence level.

6. The method of claim 1 , further comprising:

processing at least a portion of the structured contact information for the entity with a domain expert module; and

based upon the processing, generating revised structured contact information for the entity.

7. The method of claim 1 , further comprising:

providing the structure contact information for the entity and structured contact information for a plurality of other entities to a contact information database.

8. A system comprising:

a processor configured to:

receive unstructured text;

identify a plurality of text segments within the unstructured text and enriching the unstructured text by assigning one or more metadata labels to each of the text segments;

calculate one or more feature values for each text segment for at least one of the metadata labels associated with at least one of the text segments;

provide the feature values, metadata labels, and text segments to a trained machine learning system;

receive, from the machine learning system, an indication of a classification for at least one of the text segments;

identify the at least one of the text segments as a portion of contact data based upon the indication of the classification for the at least one of the text segments;

provide a confidence score for each of the classifications, wherein the at least one of the text segments is identified as a portion of contact data based upon a confidence score assigned to the classification of the at least one text segment; and

generate structured contact information for an entity based on the confidence score for each of the classifications and the text segments.

9. The system of claim 8 , further comprising a computer-based system configured to execute the machine learning system.

10. The system of claim 9 , further comprising a database storing training data to train the machine learning system.

11. The system of claim 8 , wherein the step of assigning one or more metadata labels to each of the text segments comprises:

processing each text segment of the text segments with at least one enrichment source; and

based upon an output of the at least one enrichment source, assigning the one or more metadata labels to the each text segment.

12. The system of claim 11 , wherein the at least one enrichment source is selected from the group consisting of: a dictionary of name parts; a pattern matching system; an address-matching library; and an external contact information service.

13. The system of claim 12 , wherein the at least one enrichment source comprises a plurality of different enrichment sources, each enrichment source independently selected from the group consisting of: a dictionary of name parts; a pattern matching system; an address-matching library; and an external contact information service.

14. The system of claim 8 , wherein the processor is further configured to:

process at least a portion of the structured contact information for the entity with a domain expert module; and

based upon the processing, generate revised structured contact information for the entity.

15. The system of claim 8 , wherein the processor is further configured to:

provide the structure contact information for the entity and structured contact information for a plurality of other entities to a contact information database.

16. A non-transitory computer-readable storage medium storing a plurality of instructions that, when executed, cause a processor to:

receive unstructured text;

identify a plurality of text segments within the unstructured text and enriching the unstructured text by assigning one or more metadata labels to each of the text segments;

calculate one or more feature values for each text segment for at least one of the metadata labels associated with at least one of the text segments;

provide the feature values, metadata labels, and text segments to a trained machine learning system;

receive, from the machine learning system, an indication of a classification for at least one of the text segments;

identify the at least one of the text segments as a portion of contact data based upon the indication of the classification for the at least one of the text segments;

provide a confidence score for each of the classifications, wherein the at least one of the text segments is identified as a portion of contact data based upon a confidence score assigned to the classification of the at least one text segment; and

generate structured contact information for an entity based on the confidence score for each of the classifications and the text segments.

17. The non-transitory computer-readable storage medium of claim 16 , wherein the step of assigning one or more metadata labels to each of the text segments comprises:

processing each text segment of the text segments with at least one enrichment source; and

based upon an output of the at least one enrichment source, assigning the one or more metadata labels to the each text segment.

18. The non-transitory computer-readable storage medium of claim 16 , the plurality of instructions, when executed, further causing the processor to:

process at least a portion of the structured contact information for the entity with a domain expert module; and

based upon the processing, generate revised structured contact information for the entity.

19. The non-transitory computer-readable storage medium of claim 16 , the plurality of instructions, when executed, further causing the processor to:

provide the structure contact information for the entity and structured contact information for a plurality of other entities to a contact information database.

20. A method comprising:

receiving unstructured text;

identifying a plurality of text segments within the unstructured text and enriching the unstructured text by:

processing each text segment of the plurality of text segments with at least one enrichment source; and

based upon an output of the at least one enrichment source, assigning one or more metadata labels to the each text segment;

calculating one or more feature values for each text segment;

providing the feature values, metadata labels, and text segments to a trained machine learning system;

receiving, from the machine learning system, an indication of a classification for at least one of the text segments;

based upon the indication of the classification for the at least one of the text segments, identifying the at least one of the text segments as a portion of contact data;

receiving, from the machine learning system, a confidence score for each of the classifications, wherein the at least one of the text segments is identified as a portion of contact data based upon a confidence score assigned to the classification of the at least one text segment; and

generating structured contact information for an entity based on the confidence score for each of the classifications and the text segments.

21. A system comprising:

a processor configured to:

receive unstructured text;

identify a plurality of text segments within the unstructured text and enriching the unstructured text by:

processing each text segment of the plurality of text segments with at least one enrichment source; and

based upon an output of the at least one enrichment source, assigning the one or more metadata labels to the each text segment;

calculate one or more feature values for each text segment;

provide the feature values, metadata labels, and text segments to a trained machine learning system;

receive, from the machine learning system, an indication of a classification for at least one of the text segments;

identify the at least one of the text segments as a portion of contact data based upon the indication of the classification for the at least one of the text segments;

provide a confidence score for each of the classifications, wherein the at least one of the text segments is identified as a portion of contact data based upon a confidence score assigned to the classification of the at least one text segment; and

generate structured contact information for an entity based on the confidence score for each of the classifications and the text segments.

22. A non-transitory computer-readable storage medium storing a plurality of instructions that, when executed, cause a processor to:

receive unstructured text;

identify a plurality of text segments within the unstructured text and enriching the unstructured text by:

processing each text segment of the text segments with at least one enrichment source; and

based upon an output of the at least one enrichment source, assigning the one or more metadata labels to the each text segment;

calculate one or more feature values for each text segment;

provide the feature values, metadata labels, and text segments to a trained machine learning system;

receive, from the machine learning system, an indication of a classification for at least one of the text segments;

identify the at least one of the text segments as a portion of contact data based upon the indication of the classification for the at least one of the text segments;

provide a confidence score for each of the classifications, wherein the at least one of the text segments is identified as a portion of contact data based upon a confidence score assigned to the classification of the at least one text segment; and

generate structured contact information for an entity based on the confidence score for each of the classifications and the text segments.