Semantic lexicon-based input method editor

Patent metadata

Field	Value
Publication number	US-9959340-B2
Application number	US-201213635274-A
Country	US
Kind code	B2
Filing date	Jun 29, 2012
Priority date	Jun 29, 2012
Publication date	May 1, 2018
Grant date	May 1, 2018

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.

First claim

The invention claimed is:

1. A computing system comprising:
one or more processors; and
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;
displaying a keyboard comprising a plurality of keys corresponding to input characters;
receiving, in a query window of an input method editor (IME), a query comprising one or more Latin input characters corresponding to one or more keys of the keyboard;
identifying a plurality of text candidates based on i) the query, ii) a semantic label of the of each text candidate of the plurality of text candidates in a semantic lexicon, iii) an assigned score based at least in part on the semantic label, and iv) a manifest defining a pattern to be matched by the query, each text candidate of the plurality of text candidates comprising non-Latin characters, and the semantic lexicon comprising a list of terms, information about the terms, and term relationships, wherein the assigned score is based at least in part on a matching of the query with the pattern, and wherein the semantic label acts 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;
identifying at least one rich candidate based on a first semantic label of a first text candidate of the plurality of text candidates in the semantic lexicon and based on an input into the input area of the application;
displaying, in a text candidate window of the IME, the plurality of text candidates; and
displaying, in a rich candidate window of the IME, the at least one rich candidate, wherein the at least one rich candidate includes at least one of a video and a map based on the first semantic label of the first text candidate,
wherein the query window, the rich candidate window, and text candidate window are displayed adjacent to each other in the IME.

2. The computing system as recited in claim 1 , wherein the acts further comprise:
receiving an indication of at least one text candidate or rich candidate being displayed to insert into the input area of the application; and
inserting the indicated at least one text candidate or rich candidate into the input area of the application.

3. The computing system as recited in claim 1 , wherein the at least one text candidate or rich candidate identified based on the semantic label is of a candidate type selected based on the semantic label.

4. The computing system as recited in claim 1 , wherein the pattern includes a portion prefixing or post-fixing the portion of the pattern corresponding to the semantic label.

5. The computing system of claim 1 , the operations further comprising:
collecting data related to an input scenario; and
wherein identifying the plurality of text candidates and rich candidates is further based at least in part on the input scenario.

6. One or more computer storage media maintaining instructions that, when executed by one or more processors, cause the one or more processors to perform acts comprising:
receiving, in a query window of an input method editor (IME), a query comprising one or more Latin input characters, the IME to insert a completion candidate of one or more completion candidates into an input area of an application;
assigning a semantic label to one or more portions of the query using a semantic lexicon;
determining an expected completion candidate type based at least in part on the semantic label assigned to the one or more portions of the query;
generating a plurality of text candidates of based on i) the query, ii) the assigned semantic label, iii) the semantic lexicon, iv) the expected completion candidate type, v) an assigned score based at least in part on the assigned semantic label, and vi) a manifest defining a pattern to be matched by the query, each text candidate of the plurality of text candidates comprising non-Latin characters, wherein the assigned score is based at least in part on a matching of the query with the pattern, and wherein the assigned semantic label acts as a wildcard in the pattern such that terms labeled with the assigned semantic label in the semantic lexicon match a portion of the pattern corresponding to the assigned semantic label;
generating at least one rich candidate based on a first semantic label of a first text candidate of the plurality of text candidates in the semantic lexicon and based on an input into the input area of the application;
displaying, in a text candidate window of the IME, the plurality of text candidates; and
displaying, in a rich candidate window of the IME, the at least one rich candidate, wherein the at least one rich candidate includes at least one of a video and a map based on the first semantic label of the first text candidate,
wherein the query window, the rich candidate window, and text candidate window are displayed adjacent to each other in the IME.

7. One or more computer readable media as recited in claim 6 , the acts further comprising:
receiving a selection of at least one
text candidate or rich candidate; and
inserting the at least one selected text candidate or rich candidate into the input area of the application.

8. One or more computer readable media as recited in claim 6 , wherein the determining of the semantic label using the semantic lexicon includes using a minimal perfect hash function to determine a semantic labeling of at least one portion of the query.

9. One or more computer readable media as recited in claim 6 , wherein the determining of the semantic label using the semantic lexicon includes using a fixed size bitmap indicating a semantic labeling of at least one portion of the query.

10. A method comprising:
under control of one or more processors:
receiving, in a query window of an input method editor (IME), a query comprising one or more Latin input characters corresponding to one or more keys of a keyboard;
determining a label of one or more portions of the query using a semantic lexicon comprising a list of terms, information about the terms, and term relationships;
generating a plurality of text candidates comprising non-Latin characters based on i) the determined label of the one or more portions of the query, ii) an assigned score based at least in part on the label, and iii) a manifest defining a pattern to be matched by the query, wherein the assigned score is based at least in part on a matching of the query with the pattern, and wherein the label acts as a wildcard in the pattern such that terms labeled with the label in the semantic lexicon match a portion of the pattern corresponding to the label;
generating at least one rich candidate based on a first semantic label of a first text candidate of the plurality of text candidates in the semantic lexicon and based on an input into the input area of the application;
displaying, in a text candidate window of the IME, the plurality of text candidates; and
displaying, in a rich candidate window of the IME, the at least one rich candidate, wherein the rich candidate includes at least one of a video and a map based on the first semantic label of the first text candidate,
wherein the query window, the rich candidate window, and text candidate window are displayed adjacent to each other in the IME.

11. The method as recited in claim 10 , further comprising:
receiving a selection of at least one text candidate or rich candidate; and
inserting the at least one selected text candidate or rich candidate into the input area of the application.

Assignees

• Li Chi Ho
• Xu Weijiang
• Scott Matthew Robert
• Liu Weipeng
• Chen Xi
• Chen Xi
• Microsoft Technology Licensing Llc

Inventors

• Li Chi-Ho
• Xu Weijiang
• Scott Matthew Robert
• Liu Weipeng
• Chen Xi
• Chen Xi

Classifications

• G06F16/334 Primary

  Query execution (filtering based on additional data G06F16/335) · CPC title

• G06F3/0482

  Interaction with lists of selectable items, e.g. menus · CPC title

• G06F40/30 Primary

  Semantic analysis · CPC title

• G06F17/2785

  Physics · mapped topic

• G06F17/30675 Primary

  Physics · mapped topic