US 12,481,825 B2
Method of recommending emotion icons
Kara A. Smoot, Baltimore, MD (US); Jennifer M. Hatfield, Portland, OR (US); Kelley Danyi, Virginia Beach, VA (US); Jill S. Dhillon, Jupiter, FL (US); Juhi Bharat, Highland Park, NJ (US); Vinod A. Valecha, Pune (IN); and Uma Subbu, Chicago, IL (US)
Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed by INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed on Sep. 6, 2023, as Appl. No. 18/461,590.
Prior Publication US 2025/0077768 A1, Mar. 6, 2025
Int. Cl. G06F 40/20 (2020.01)
CPC G06F 40/20 (2020.01) 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for recommending emotion icons, the computer-implemented method comprising:
detecting text associated with user activity within a connected messaging environment;
in response to detecting the text associated with the user activity within the connected messaging environment, determining a context for the detected text based on at least location data, environment data, language context data, and historical data,
wherein determining the context based on the location data further comprises determining a location of a device associated with the connected messaging environment, and determining emotion icons based on a location type and the location data, wherein the location data further comprises language characteristics of a given location,
wherein determining the context based on the environment data further comprises determining whether a detected location marked within at least one of the device and the connected messaging environment is labeled with an environment, and determining emotion icons based on the environment,
wherein determining the context based on the language context data further comprises using machine learning to determine language context of the detected text, wherein the machine learning is trained on a dataset of messages and emotion icon usage associated with each message to learn the context of the dataset of messages based on language processing, and
wherein determining the context based on the historical data further comprises identifying emotion icons a user selected for the detected text and similar text in previous messaging;
generating a filtered series of emotion icons based on the determined context; and
displaying the filtered series of emotion icons to a user for selection.