Phrase-based data classification system

• US 9,043,285 B2
• Filed: 02/09/2012
• Issued: 05/26/2015
• Est. Priority Date: 10/28/2011
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: LinkedIn Corporation (Microsoft Corporation)
• Inventors: Bekkerman, Ron
• Primary Examiner(s): AL HASHEMI, SANA A
• Application Number:
  US13/369,912
• Publication Number:
  US 20130110498A1
• Time in Patent Office: 1,202 Days
• Field of Search: 707/609, 707/661, 707/687, 707/813, 707/821, 707/953
• US Class Current: 707/661
• CPC Class Codes: G06F 40/205 Parsing
• CPC Class Codes: G06F 40/258 Heading extraction; Automatic titling; Numbering
• CPC Class Codes: G06F 40/268 Morphological analysis
• CPC Class Codes: G06F 40/289 Phrasal analysis, e.g. finite state techniques or chunking
• CPC Class Codes: G06Q 10/1053 Employment or hiring
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: Y10S 707/953 Organization of data

Abstract

A method of classifying data is disclosed. Text data items are received. A set of classes into which the text data items are to be classified is received. A phrase-based classifier to classify the text data items into the set of classes is selected. The phrase-based classifier is applied to classify the text data items into the classes. Here, the applying includes creating a controlled vocabulary pertaining to classifying the text data items into the set of classes, building phrases based on the text data items and the controlled vocabulary, and classifying the text data items into the set of classes based on the phrases.

Claims

1. A method comprising:

receiving text data items;

receiving a set of classes into which the text data items are to be classified;

selecting a phrase-based classifier to classify the text data items into the set of classes; and

applying the phrase-based classifier to classify the text data items into the classes, the applying including;

creating a controlled vocabulary pertaining to classifying the text data items into the set of classes;

building phrases based on the text data items and the controlled vocabulary;

classifying, using at least one processor, the text data items into the set of classes based on the phrases; and

reclassifying a text data item of the text data items into the set of classes based on a consistency error.

4. The method of claim 1, the applying further comprising reclassifying a text data item of the text data items into the set of classes based on a systematic error.

5. The method of claim 1, the applying further comprising evaluating a precision of the classifying of the text data items.

6. The method of claim 1, the selecting of the phrase-based classifier being based on the text data items being self-descriptive of an attribute of the users of the social networking site.

12. The method of claim 1, wherein the attribute is a job title.

2. A system comprising:

at least one processor; and

a data classification system implemented by the at least one processor and including;

an online module configured to;

receive text data items;

receive a set of classes into which the text data items are to be classified;

select a phrase-based classifier to classify the text data items into the set of classes; and

apply the phrase-based classifier to classify the text data items into the classes, the applying including;

receiving a set of classes into which text data items are to be classified;

creating a controlled vocabulary pertaining to classifying the text data items into the set of classes;

building phrases based on the text data items and the controlled vocabulary; and

classifying the text data items into the set of classes based on the phrases; and

reclassifying a text data item of the text data items into the set of classes based on a consistency error.

7. The system of claim 2, the applying further comprising reclassifying a text data item of the text data items into the set of classes based on a systematic error.

8. The system of claim 2, further comprising evaluating a precision of the classifying of the text data items.

9. The system of claim 2, wherein the online module is configured to select the phrase-based classifier based on the text data items being self-descriptive of an attribute of the users of the social networking site.

13. The system of claim 9, wherein the attribute is a job title.

3. A non-transitory machine readable medium embodying a set of instructions that, when executed by a processor, cause the processor to perform a method, the method comprising:

receiving text data items;

receiving a set of classes into which the text data items are to be classified;

selecting a phrase-based classifier to classify the text data items into the set of classes; and

applying the phrase-based classifier to classify the text data items into the classes, the applying including;

creating a controlled vocabulary pertaining to classifying the text data items into the set of classes;

building phrases based on the text data items and the controlled vocabulary; and

classifying, using at least one processor, the text data items into the set of classes based on the phrases; and

reclassifying a text data item of the text data items into the set of classes based on a consistency error.

10. The non-transitory machine readable medium of claim 3, applying further comprising reclassifying a text data item of the text data items into the set of classes based on a systematic error.

11. The non-transitory machine readable medium of claim 3, applying further comprising evaluating a precision of the classifying of the text data items.

14. The non-transitory machine readable medium of claim 3, the selecting of the phrase-based classifier being based on the text data items being self-descriptive of an attribute of the users of the social networking site.