Semantic Lexicon-Based Input Method Editor

• US 20150121290A1
• Filed: 06/29/2012
• Published: 04/30/2015
• Est. Priority Date: 06/29/2012
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Li, Chi-Ho, Xu, Weijiang, Scott, Matthew Robert, Liu, Weipeng, Chen, Xi
• Granted Patent: US 9,959,340 B2
• Time in Patent Office: Days
• US Class Current: 715/780
• CPC Class Codes: G06F 16/334 Query execution G06F16/335 takes precedence
• CPC Class Codes: G06F 3/0482 Interaction with lists of selectable items, e.g. menus
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

Some implementations provide techniques and arrangements for semantic lexicon based processing, such as in an input method editor. In some instances, a semantic label may be received that is to be defined for a semantic lexicon and at least a first term may be identified as a positive or negative example of the semantic label. In response, some examples may label at least a second term in the semantic lexicon with the semantic label based at least in part on the identification of the first term as a positive or negative example of the semantic label.

Claims

1. A computing system comprising:

one or more processors;

one or more computer readable media maintaining instructions that, when executed by the one or more processors, cause the one or more processors to perform acts comprising;

displaying a user interface of an application including an input area;

receiving, by an input method editor (IME), a query, the IME to insert a completion candidate of one or more completion candidates into the input area of the application; and

displaying the one or more completion candidates, at least one of the completion candidates being selected for display based on a semantic label.

2. The computing system as recited in claim 1, further comprising:

receiving an indication of at least one of the one or more displayed completion candidates to insert into the input area of the application; and

inserting the at least one indicated completion candidate into the input area of the application.

3. The computing system as recited in claim 1, wherein a portion of the query is labeled with the semantic label in a semantic lexicon.

4. The computing system as recited in claim 3, wherein the at least one completion candidate selected for display based on the semantic label is of a candidate type selected based on the semantic label.

5. The computing system as recited in claim 1, wherein the one or more completion candidates displayed are selected based on a respective score assigned to each completion candidate, the score being assigned to at least one completion candidate based at least in part on the semantic label.

6. The computing system as recited in claim 5, wherein the score assigned to the at least one completion candidate selected for display based on the semantic label is based at least in part on a matching of the query with a pattern, the semantic label acting as a wildcard in the pattern such that terms labeled with the semantic label in the semantic lexicon match a portion of the pattern corresponding to the semantic label.

7. The computing system as recited in claim 6, wherein the pattern includes a portion prefixing or post-fixing the portion of the pattern corresponding to the semantic label.

8. The computing system as recited in claim 4, wherein at least one of the completion candidates selected for display based on the semantic label is a rich candidate.

9. A method comprising:

receiving a semantic label to be defined for a semantic lexicon and at least a first term identified as a positive or negative example of the semantic label; and

labeling one or more second terms in the semantic lexicon with the semantic label based at least in part on the identification of the first term as a positive or negative example of the semantic label.

10. A method as recited in claim 9, wherein the labeling of the second term is based at least in part on one or more term relationships between the second term and at least the first term.

11. A method as recited in claim 9, further comprising:

identifying at least one third term in the semantic lexicon that is a marginal term not labeled by the labeling;

receiving a label of the at least one third term as a positive or negative example of the semantic label; and

labeling on or more fourth terms in the semantic lexicon with the semantic label based at least in part on the labeling of the third term.

12. A method as recited in claim 11, wherein the acts of identifying, receiving, and labeling one or more fourth terms are repeated until a condition is met.

13. A method as recited in claim 12, wherein the condition is one of a number of marginal terms or a measure of marginal terms.

14. A method as recited in claim 11, wherein:

the identifying includes determining a tentative labeling of the third term with regard to the semantic label; and

the receiving of the label of the at least one third term confirms or negates the tentative labeling.

15. A method as recited in claim 14, wherein the acts of identifying, receiving, and labeling one or more fourth terms are repeated until a condition regarding the confirmation or negation of the tentative labeling of one or more third terms is met.

16. One or more computer readable media maintaining instructions that, when executed by one or more processors, cause the one or more processors to perform acts comprising:

receiving, by an input method editor (IME), a query, the IME to insert a completion candidate of one or more completion candidates into the input area of an application;

determining a semantic labeling of one or more portions of the query using a semantic lexicon;

generating one or more completion candidates based at least in part on the determined semantic labeling; and

displaying the one or more completion candidates.

17. One or more computer readable media as recited in claim 16, the acts further comprising:

receiving a selection of at least one of the one or more displayed completion candidates; and

inserting the at least one selected completion candidate into the input area of the application.

18. One or more computer readable media as recited in claim 16, wherein the determining of the semantic labeling using the semantic lexicon includes using a minimal perfect hash function to determine the semantic labeling of at least one portion of the query.

19. One or more computer readable media as recited in claim 16, wherein the determining of the semantic labeling using the semantic lexicon includes using a fixed size bitmap indicating the semantic labeling of at least one portion of the query.

20. One or more computer readable media as recited in claim 16, wherein the generating one or more completion candidates includes determining one or more types of completion candidates to generate based on a manifest defining a pattern to be matched by the query, the semantic label acting as a wildcard in the pattern such that terms labeled with the semantic label in the semantic lexicon match a portion of the pattern corresponding to the semantic label.