Semantic content searching

• US 8,380,719 B2
• Filed: 06/18/2010
• Issued: 02/19/2013
• Est. Priority Date: 06/18/2010
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Chang, Eric I-Chao, Gillam, Michael Todd, Xu, Yan, Feied, Craig, Handler, Jonathan
• Primary Examiner(s): OBISESAN, AUGUSTINE KUNLE
• Application Number:
  US12/818,286
• Publication Number:
  US 20110314024A1
• Time in Patent Office: 977 Days
• Field of Search: None
• US Class Current: 707/740
• CPC Class Codes: G06F 16/3325 Reformulation based on results of preceding query

Abstract

One or more techniques and/or systems are disclosed that provide for document retrieval where a user can identify key attributes of potential target documents that are desirable (e.g., have a particular semantic content for the user). Further, relevant documents that comprise the desired semantic content can be retrieved. Additionally, the user can provide feedback on the retrieved documents, for example, based on key semantic concepts found in the documents, and the input can be used to update the classification. For example, this process can be iterated to improve the retrieval and precision of documents found through machine learning techniques.

Claims

1. A computer-based method for document searching by semantic content, comprising:

receiving a user selection of a first portion of a document, the first portion comprising desired semantic content;

running the document, comprising the selected first portion, through one or more classifiers to identify a first potential target document comprising a second portion;

responsive to determining that the second portion of the first potential target document does not have the desired semantic content;

receiving a user selection of a third portion of the first potential target document that comprises the desired semantic content; and

running the first potential target document, comprising the selected third portion, through the one or more classifiers to identify a second potential target document comprising a fourth portion that has the desired semantic content; and

at least one of;

receiving user input comprising an indication that a second document identified by the one or more classifiers comprises the desired semantic content;

receiving user input comprising an indication that the second document identified by the one or more classifiers does not comprise the desired semantic content;

receiving user input comprising a selected fifth portion of the second document, the selected fifth portion comprising the desired semantic content;

running a plurality of documents through the one or more classifiers until a desired document selection accuracy is reached for the desired semantic content;

running the plurality of documents through the one or more classifiers until a desired number of correct documents are retrieved without an incorrect document being retrieved;

running the plurality of documents through the one or more classifiers until a desired number of documents have been retrieved;

using a second classifier to validate a document retrieved by a first classifier;

or

identifying a combination of two or more classifiers that has a desired accuracy rate for retrieving documents, and utilizing the identified combination to retrieve documents for the desired semantic content.

2. The method of claim 1, comprising identifying one or more documents by performing a keyword search on a collection of documents.

3. The method of claim 1, comprising responsive to determining that the second portion of the first potential target document has the desired semantic content, running the one or more classifiers to select a third potential target document.

4. The method of claim 1, the document comprising metadata labels that describe one or more attributes of one or more components of the document.

5. The method of claim 1, comprising utilizing user input to update the one or more classifiers.

6. The method of claim 1, comprising running the plurality of documents through the one or more classifiers until a desired threshold is reached.

7. The method of claim 1, the second document comprising at least some of at least one of the document, the first potential target document, or the second potential target document.

8. The method of claim 1, the one or more classifiers comprising one or more trained classifiers.

9. The method of claim 1, comprising presenting a potential target document to a user merely when document classification provides a result that is within a desired threshold.

10. The method of claim 9, the desired threshold comprising an uncertainty threshold for a classifier model.

11. The method of claim 1, comprising utilizing a plurality of classifiers concurrently to retrieve one or more potential target documents.

12. The method of claim 11, comprising:

identifying a third classifier that has a desired accuracy rate for retrieving potential target documents.

13. The method of claim 12, comprising utilizing the identified third classifier to retrieve one or more documents.

14. A system, comprising:

one or more processing units; and

memory comprising instructions that when executed by at least some of the one or more processing units, perform a method comprising;

receiving a user selection of a first portion of a document, the first portion comprising desired semantic content;

running the document, comprising the selected first portion, through one or more classifiers to identify a first potential target document comprising a second portion;

responsive to determining that the second portion of the first potential target document does not have the desired semantic content;

receiving a user selection of a third portion of the first potential target document that comprises the desired semantic content; and

running the first potential target document, comprising the selected third portion, through the one or more classifiers to identify a second potential target document comprising a fourth portion that has the desired semantic content; and

at least one of;

receiving user input comprising an indication that a second document identified by the one or more classifiers comprises the desired semantic content;

receiving user input comprising an indication that the second document identified by the one or more classifiers does not comprise the desired semantic content;

receiving user input comprising a selected fifth portion of the second document, the selected fifth portion comprising the desired semantic content;

running a plurality of documents through the one or more classifiers until a desired document selection accuracy is reached for the desired semantic content;

running the plurality of documents through the one or more classifiers until a desired number of correct documents are retrieved without an incorrect document being retrieved;

running the plurality of documents through the one or more classifiers until a desired number of documents have been retrieved;

using a second classifier to validate a document retrieved by a first classifier;

or

identifying a combination of two or more classifiers that has a desired accuracy rate for retrieving documents, and utilizing the identified combination to retrieve documents for the desired semantic content.

15. The system of claim 14, the desired semantic content associated with at least one of a social security number or a telephone number.

16. The system of claim 14, the second document comprising at least some of at least one of the document, the first potential target document, or the second potential target document.

17. A non-signal computer readable storage device comprising instructions that when executed, perform a method comprising:

receiving a user selection of a first portion of a document, the first portion comprising desired semantic content;

running the document, comprising the selected first portion, through one or more classifiers to identify a first potential target document comprising a second portion;

responsive to determining that the second portion of the first potential target document does not have the desired semantic content;

receiving a user selection of a third portion of the first potential target document that comprises the desired semantic content; and

running the first potential target document, comprising the selected third portion, through the one or more classifiers to identify a second potential target document comprising a fourth portion that has the desired semantic content; and

at least one of;

receiving user input comprising an indication that a second document identified by the one or more classifiers comprises the desired semantic content;

receiving user input comprising an indication that the second document identified by the one or more classifiers does not comprise the desired semantic content;

receiving user input comprising a selected fifth portion of the second document, the selected fifth portion comprising the desired semantic content;

running a plurality of documents through the one or more classifiers until a desired document selection accuracy is reached for the desired semantic content;

running the plurality of documents through the one or more classifiers until a desired number of correct documents are retrieved without an incorrect document being retrieved;

running the plurality of documents through the one or more classifiers until a desired number of documents have been retrieved;

using a second classifier to validate a document retrieved by a first classifier;

or

identifying a combination of two or more classifiers that has a desired accuracy rate for retrieving documents, and utilizing the identified combination to retrieve documents for the desired semantic content.

18. The non-signal computer readable storage device of claim 17, the desired semantic content associated with at least one of a social security number or a telephone number.

19. The non-signal computer readable storage device of claim 17, the second document comprising at least some of at least one of the document, the first potential target document, or the second potential target document.

20. The non-signal computer readable storage device of claim 17, the document comprising metadata labels that describe one or more attributes of one or more components of the document.