# US 10,372,310 B2 · Suppression of input images > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,372,310 B2 - Document ID: US10372310B2 - Status: Granted - Kind: B2 - Application number: 15/257,633 - Filed: 2016-09-06 - Published: 2017-12-28 - Granted: 2019-08-06 - Earliest claimed: 2016-06-23 - Assignees: Microsoft Technology Licensing, LLC - Applicants: Microsoft Technology Licensing, LLC - Inventors: David Martinez Del Corral; Alice Elizabeth Rosam; Vladyslav Kolesnyk - Web page: https://patentgrep.com/patent/US10372310B2 ## CPC - G06F 3/04842: Selection of displayed objects or displayed text elements (G06F3/0482 takes precedence) - G06F 3/0236: {using selection techniques to select from displayed items} - G06F 3/0237: {using prediction or retrieval techniques} - G06F 3/04883: for inputting data by handwriting, e.g. gesture or text - G06F 3/04886: by partitioning the display area of the touch-screen or the surface of the digitising tablet into independently controllable areas, e.g. virtual keyboards or menus - G06F 40/232: Orthographic correction, e.g. spell checking or vowelisation - G06F 40/274: Converting codes to words; Guess-ahead of partial word inputs - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract A computing device is described which has a memory storing at least one indicator of image use; a user interface which receives user input; and a processor configured to trigger prediction, from the user input, of a plurality of candidate images for input to the computing device. The processor is configured to at least partially suppress the prediction of the plurality of images using the indicator of image use. ## Claims ### Claim 1 A computing device comprising: a user interface which receives user input; a memory storing context-specific indicators of image use, where image use is entry of images to the computing device, each of the indicators of image use being a statistical representation of observed image use measuring how likely a user is to input an image into the user interface, the context-specific indicators being a combination of one or more of application-specific, user-specific, field-specific, enterprise-specific, recipient-specific, user-age-group-specific, user-gender-specific, topic-specific, population-specific, communication-specific, language-specific, geographical-region-specific, time-zone-specific, or time-of-year-specific indicators; a processor implementing a prediction tool with capability to predict, from the user input, candidate text and candidate images for input to the computing device, wherein the prediction tool is configured to select, based on context associated with the user input, at least one of the context-specific indicators of image use, and, if the selected at least one indicator of image use is below a threshold associated with rare usage, to automatically, without manual input from the user for adjusting prediction functionality of the prediction tool, suppress candidate image predictions available for selection by a user. ### Claim 2 (depends on 1) A computing device as claimed in claim 1 wherein the processor is configured to dynamically compute at least one of the indicators of image use during operation of the computing device. ### Claim 3 (depends on 1) A computing device as claimed in claim 1 wherein the prediction tool is configured to suppress the candidate image predictions by switching off the image prediction capability. ### Claim 4 (depends on 1) A computing device as claimed in claim 1 wherein the indicators comprise a ratio of observations of images input to the computing device to observations of candidate images available for selection by a user. ### Claim 5 (depends on 1) A computing device as claimed in claim 1 wherein the user interface is a soft keyboard having text prediction and image prediction capability. ### Claim 6 A computer-implemented method comprising: storing context-specific indicators of image use, where image use is entry of images to a computing device, each of the indicators of image use being a statistical representations of observed image use measuring how likely a user is to input an image into the user interface, the context-specific indicators being a combination of one or more of application-specific, user-specific, field-specific, enterprise-specific, recipient-specific, user-age-group-specific, user-gender-specific, topic-specific, population-specific, communication-specific, language-specific, geographical-region-specific, time-zone-specific, or time-of-year-specific; receiving user input; selecting, based on context associated with the user input, at least one of the context-specific indicators of image use; in response to the selected at least one indicator of image use being at or above a threshold associated with rare usage, triggering prediction, from the user input, of at least one of candidate text or a plurality of candidate images for input to the computing device; and in response to the selected at least one indicator of image use being below the threshold associated with rare usage, automatically, without manual input from the user for adjusting prediction functionality, suppressing candidate image predictions available for selection by a user. ### Claim 7 (depends on 6) A method as claimed in claim 6, comprising dynamically computing at least one of the indicators of image use during operation of the computing device. ### Claim 8 A computing device comprising: a processor which implements a soft keyboard having text prediction and emoji prediction capability and providing a user interface which receives user input; and a memory storing a plurality of applications executable on the computing device, the memory storing an application-specific indicator of emoji use for each of the applications, the application-specific indicators for the plurality of application being statistical representations of observed emoji use measuring how likely a user is to input an emoji into the user interface; and wherein the processor is configured to select, based on an application in focus when the user input is received, at least one of the application-specific indicators of emoji use, and, if the selected at least one indicator of emoji use is below a threshold associated with rare usage, to automatically switch off the emoji prediction capability of the keyboard.