Enrichment of named entities in documents via contextual attribute ranking

• US 9,552,352 B2
• Filed: 11/10/2011
• Issued: 01/24/2017
• Est. Priority Date: 11/10/2011
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Viegas, Evelyne, Rozell, Eric Anthony
• Primary Examiner(s): Kells, Asher
• Application Number:
  US13/293,139
• Publication Number:
  US 20130124964A1
• Time in Patent Office: 1,902 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 40/295 Named entity recognition

Abstract

Technologies pertaining to retrieval of contextually relevant attribute values for an automatically identified named entity in a document are described herein. Named entity recognition technologies are employed to identify named entities in the text of a document. Context corresponding to an identified named entity is analyzed to probabilistically assign a class to the named entity. Attributes that are most relevant to the class are determined, and attribute values for such attributes are retrieved. The attribute values are presented in correlation with the named entity in the document responsive to user-selection of the named entity in the document.

Claims

1. A method comprising plurality of acts that are executed by at least one processor in at least one computing device, the acts comprising:

extracting text from a document;

detecting existence of a named entity in the text of the document;

generating annotated text responsive to detecting the existence of the named entity in the text of the document, the annotated text comprising an annotation that identifies the named entity in the text;

transmitting the annotated text to an entity classification system, the entity classification system assigning a class from a plurality of possible classes to the named entity based at least in part upon the annotated text;

receiving from the entity classification system data that identifies the class assigned to the named entity by the entity classification system;

transmitting the data that identifies the class to an attribute ranking system, the attribute ranking system determining a plurality of attributes that are relevant to the class;

receiving from the attribute ranking system the plurality of attributes that are relevant to the class;

responsive to receiving the plurality of attributes, retrieving values for the plurality of attributes for the named entity;

subsequent to retrieving the values for the plurality of attributes, causing the named entity to be highlighted in the document when the document is displayed on a display of a client computing device, the named entity highlighted to inform a user of the client computing device that additional data about the named entity is available, wherein the client computing device displays, on the display, the values for the plurality of attributes for the named entity responsive to detecting that the user has selected the named entity in the document.

2. The method of claim 1, wherein detecting existence of the named entity in the text of the document comprises executing at least one natural language processing algorithm to detect the existence of the named entity in the text of the document.

3. The method of claim 1, wherein the document comprises metadata that identifies the named entity in the document, and wherein detecting existence of the named entity in the text of the document comprises analyzing the metadata.

4. The method of claim 1, wherein the document is a web page, and wherein the values for the plurality of attributes for the named entity are displayed in the web page without causing the user to leave the web page.

5. The method of claim 4 being executed in a web browser.

6. The method of claim 1, wherein the document is a word processing document, the method further comprising performing the acts of the method while the user modifies the word processing document.

7. The method of claim 1, wherein the client computing device displays, on the display, the values for the plurality of attributes for the named entity responsive to detecting that the user has one of hovered a cursor over the named entity, made a gesture towards the named entity, selected the named entity via touch, or eyes of the user being focused on the named entity.

8. The method of claim 1, wherein retrieving values for the plurality of attributes for the named entity comprises:

constructing a query that comprises the named entity and at least one attribute from the plurality of attributes; and

searching over a web-accessible data repository through utilization of the query.

9. The method of claim 8, wherein the web-accessible data repository comprises structured data.

10. The method of claim 1, further comprising:

disambiguating the text in the document to identify a plurality of different named entities;

acquiring attribute values corresponding to attributes of each named entity identified in the text; and

causing portions of the text to be highlighted to inform the user that the attribute values are available with respect to each named entity identified in the text.

11. The method of claim 1, further comprising:

accessing a data repository that comprises user history data; and

at least one of transmitting the user history data to the entity classification system, wherein the entity classification system assigns the class to the named entity based at least in part upon the user history data;

or

transmitting the user history data to the attribute ranking system, wherein the attribute ranking system determines the plurality of attributes that are relevant to the class based at least in part upon the user history data.

12. A system that facilitates provision of contextually relevant data to a user, the system comprising:

at least one processor; and

memory comprising instructions that, when executed by the at least one processor, cause the at least one processor to perform acts comprising;

receiving a document that comprises text, the text comprising a named entity;

identifying the named entity in the text;

responsive to identifying the named entity in the text and based upon the named entity, inferring a class to which the named entity belongs, the class being amongst a plurality of possible classes;

responsive to inferring the class to which the named entity belongs transmitting the data that identifies the class to an attribute ranking system, the attribute ranking system determining attributes that are relevant to the class;

receiving from the attribute ranking system the attributes that are relevant to the class; and

retrieving values for the attributes for the named entity responsive to receiving the attributes;

wherein the values for the attributes for the named entity are displayed as a graphical overlay in the document responsive to a reviewer of the document selecting the named entity on a client computing device.

13. The system of claim 12, wherein the named entity is highlighted in the document to inform the reviewer of the document that the values for the attributes have been retrieved for the named entity.

14. The system of claim 12, the at least one processor executes the instructions in response to a web browser requesting the document.

15. The system of claim 12, the at least one processor executes the instructions when executing a word processing application.

16. The system of claim 12, wherein the acts further comprise receiving an indication from the client computing device that the reviewer of the document has selected the named entity in the document.

17. The system of claim 12, wherein a tooltip that comprises the attribute values is displayed on the client computing device responsive to the reviewer of the document selecting the named entity at the client computing device.

18. The system of claim 12, wherein the document is a word processing document, wherein retrieving the attribute values comprises retrieving the attribute values as the reviewer of the word processing document is modifying the word processing document, wherein named entities entered into the word processing document are highlighted immediately responsive to retrieving the attribute values.

20. The system of claim 12, wherein a pop-up window that comprises the attribute values is displayed on the client computing device responsive to the reviewer of the document selecting the named entity at the client computing device.

19. A computer-readable medium comprising instructions that, when executed by a processor, cause the processor to perform acts comprising:

identifying named entities in text of a web page;

subsequent to identifying the named entities in the text, annotating the named entities in the text to form annotated text;

transmitting the annotated text to a named entity classification system, wherein the named entity classification system probabilistically assigns classes to respective named entities in the annotated text;

receiving class assignments for the respective named entities from the named entity classification system;

for each unique class received from the named entity classification system, transmitting the respective unique class to an attribute ranking system, wherein the attribute ranking system outputs lists of attributes that are relevant to the respective classes;

receiving from the attribute ranking system, for each unique class, attributes that are relevant to a respective unique class;

retrieving attribute values for the received attributes for each named entity in the text of the web page, wherein the attribute values are retrieved from a data repository that comprises structured data, wherein each named entity in the web page has at least one attribute value associated therewith to inform a reviewer of the web page that additional information pertaining to the named entities is available, wherein attribute identities and corresponding values for a named entity are displayed in response to the named entity being selected in the web page.