TOOLS AND TECHNIQUES FOR EXTRACTING KNOWLEDGE FROM UNSTRUCTURED DATA RETRIEVED FROM PERSONAL DATA SOURCES

• US 20130110842A1
• Filed: 11/02/2011
• Published: 05/02/2013
• Est. Priority Date: 11/02/2011
• Status: Active Grant

• Current Assignee: Salesforce.com, Inc.
• Original Assignee: SRI International, Inc.
• Inventors: Donneau-Golencer, Thierry, Yarlagadda, Madhu, Nitz, Kenneth C., Mark, William Scott, Hulen, Corey, Singh, Rajan
• Granted Patent: US 9,443,007 B2
• Time in Patent Office: Days
• US Class Current: 707/741
• CPC Class Codes: G06F 16/22 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: G06F 16/24522 Translation of natural language queries to structured queries
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 16/337 Profile generation, learning or modification

Abstract

A system may include multiple personal data sources and a machine-implemented data extractor and correlator configured to retrieve personal data from at least one of the personal data sources. The data extractor and correlator may extract information from unstructured data within the retrieved personal data and correlate the extracted information with previously stored structured data to generate additional structured data. The system may also include a storage device configured to store the previously stored structured data and the additional structured data. A natural language query module may be configured to receive a natural language query from a user and provide a response to the natural language query based at least in part on one or both of the previously stored structured data and the additional structured data.

Claims

1. A system, comprising:

a plurality of personal data sources;

a machine-implemented data extractor and correlator configured to retrieve personal data from at least one of the plurality of personal data sources, extract information from unstructured data within the retrieved personal data, and correlate the extracted information with previously stored structured data to generate additional structured data;

a storage device configured to store the previously stored structured data and the additional structured data; and

a natural language query module configured to receive a natural language query from a user and provide a response to the natural language query based at least in part on one or both of the previously stored structured data and the additional structured data.

2. The system of claim 1, wherein the plurality of personal data sources comprises at least two of a group consisting of:

an email message, a calendar item, a customer relationship management (CRM) application object, an address book entry, a tweet, a blog entry, a file, a folder, a presentation, and a document.

3. The system of claim 1, wherein the retrieved personal data comprises supplemental structured data.

4. The system of claim 3, wherein the data extractor and correlator is further configured to correlate the supplemental structured data with one or both of the previously stored structured data and the additional structured data to generate further structured data.

5. The system of claim 1, wherein the data extractor and correlator comprises a semantic analysis module configured to determine semantic information based on the unstructured data, and wherein the extracted information comprises the semantic information.

6. The system of claim 5, wherein the data extractor and correlator is configured to correlate the extracted information with the previously stored structured data based at least in part on the semantic information.

7. The system of claim 1, wherein the data extractor and correlator comprises a semantic analysis module configured to determine semantic information based on public data retrieved from at least one public data source, and wherein the data extractor and correlator is further configured to correlate the semantic information with one or both of the previously stored structured data and the additional structured data to generate further structured data.

8. The system of claim 1, further comprising a machine-implemented user profile module configured to update a user profile based at least in part on one or both of the previously stored structured data and the additional structured data.

9. The system of claim 8, wherein the user profile module is further configured to generate a new user profile based at least in part on one or both of the previously stored structured data and the additional structured data.

10. The system of claim 8, further comprising at least one user application configured to provide user interaction based at least in part on the user profile.

11. The system of claim 8, further comprising a feedback module configured to update the user profile responsive to one or both of the previously stored structured data and the additional structured data.

12. The system of claim 11, wherein the feedback module is further configured to update the user profile responsive to public data retrieved from at least one public data source.

13. The system of claim 1, further comprising a machine-implemented document harvester configured to retrieve a document from at least one of the plurality of personal data sources.

14. The system of claim 13, further comprising a machine-implemented document indexer configured to index a plurality of documents harvested by the document harvester.

15. The system of claim 1, wherein the machine-implemented data extractor and correlator is further configured to retrieve public data from at least one public data source, extract public information from unstructured data within the retrieved public data, and correlate the extracted public information with the previously stored structured data to generate further additional structured data.

16. The system of claim 1, wherein the data extractor and correlator is configured to extract information from the unstructured data by performing at least one of a group consisting of:

breaking at least one sentence into subject, verb, and object (SVO);

extracting phrases that link a subject to an object;

extracting at least one word in close proximity to an identified feature; and

extracting at least one word in close proximity to a known quality.

17. The system of claim 1, wherein the data extractor and correlator is configured to correlate the extracted information with the previously stored structured data by building at least one relationship between the extracted information and the previously stored structured data.

18. The system of claim 17, wherein the data extractor and correlator is configured to build the at least one relationship between the extracted information and the previously stored structured data by tagging a feature identified in the extracted information.

19. A machine-implemented method, comprising:

a device retrieving personal data from at least one of a plurality of personal data sources;

a processor extracting information from unstructured data within the retrieved personal data;

the processor correlating the extracted information with previously stored structured data to generate additional structured data;

a storage device storing the previously stored structured data and the additional structured data;

the device receiving a natural language query from a user; and

the processor determining a response to the natural language query to be provided to the user, wherein the response is based at least in part on one or both of the previously stored structured data and the additional structured data.

20. The machine-implemented method of claim 19, further comprising:

generating a user profile based at least in part on one or both of the previously stored structured data and the additional structured data.

21. The machine-implemented method of claim 19, further comprising:

updating a previously generated user profile based at least in part on one or both of the previously stored structured data and the additional structured data.

22. The machine-implemented method of claim 19, further comprising.

the device retrieving public data from at least one of a plurality of public data sources;

the processor extracting public information from unstructured data within the retrieved public data; and

the processor correlating the extracted public information with the previously stored structured data to generate further additional structured data.

23. The machine-implemented method of claim 22, further comprising:

updating a previously generated user profile based at least in part on the further additional structured data.