Systems and methods for classifying electronic documents

• US 9,710,540 B2
• Filed: 03/14/2016
• Issued: 07/18/2017
• Est. Priority Date: 03/15/2013
• Status: Active Grant

• Current Assignee: Tsg Technologies, LLC
• Original Assignee: Tsg Technologies, LLC
• Inventors: Lu, Caimei, Prosser, Ricky John, Bryars, Michael Wayne
• Primary Examiner(s): NGUYEN, CAM LINH T
• Application Number:
  US15/069,661
• Publication Number:
  US 20160267169A1
• Time in Patent Office: 491 Days
• Field of Search: 707738
• CPC Class Codes: G06F 16/245 Query processing
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/353 into predefined classes
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

A method of classifying an electronic document from an electronic media source includes generating statistical data from one or more training documents and creating a plurality of classification rules, including creating at least one topic model-based classification rule using the statistical data and creating at least one query-based classification rule using one or more user defined categories and the statistical data. The method further includes classifying the electronic document using the at least one topic model-based classification rule. Example systems for classifying an electronic document from an electronic media source are also disclosed.

Claims

1. A method of classifying an electronic document from an electronic media source, the method comprising:

generating statistical data from one or more training documents,

creating a plurality of classification rules, including creating at least one topic model-based classification rule using the statistical data, and creating at least one query-based classification rule using one or more user defined categories and the statistical data, and

classifying the electronic document using the at least one topic model-based classification rule.

2. The method of claim 1 wherein classifying includes classifying the electronic document using the at least one topic model-based classification rule and the at least one query-based classification rule.

3. The method of claim 1 further comprising filtering each training document prior to creating the classification rules.

4. The method of claim 1 wherein creating includes evaluating a classification rule of the classification rules on a test document to ensure the classification rule meets a defined threshold.

5. The method of claim 4 wherein creating includes modifying the classification rule if the classification rule is below the defined threshold.

6. The method of claim 1 wherein the statistical data includes a topic-term matrix and a document-topic matrix.

7. The method of claim 6 wherein the topic-term matrix includes one or more topical categories and wherein classifying includes assigning at least one of the one or more topical categories of the topic-term matrix to the electronic document using the at least one topic model-based classification rule.

8. The method of claim 7 further comprising outputting the assigned at least one topical category of the electronic document.

9. The method of claim 6 wherein creating includes creating the at least one query-based classification rule using the topic-term matrix.

10. The method of claim 9 further comprising selecting a query-based classification rule of the at least one query-based classification rule based on the assigned at least one topical category and wherein classifying includes categorizing the electronic document into at least one user defined category of the one or more user defined categories using the selected query-based classification rule.

11. The method of claim 10 further comprising outputting the at least one user defined category of the one or more user defined categories.

12. The method of claim 1 wherein classifying includes classifying the electronic document only using the at least one topic model-based classification rule.

13. The method of claim 1 further comprising modifying the plurality of classification rules after a period of time.

14. The method of claim 1 wherein the one or more training documents include one or more unlabeled training documents.

15. A method of classifying an electronic document from an electronic media source using at least one classification rule from a plurality of classification rules, the plurality of classification rules including at least one topic model-based classification rule and at least one query-based classification rule, the topic model-based classification rule created using statistical data generated from one or more training documents, the query-based classification rule created using the statistical data and one or more user defined categories, the method comprising classifying the electronic document using the at least one topic model-based classification rule.