Granted Patent B2

US 12,592,310 · App. 17/475,839 · Granted Mar 31, 2026

Inventors: Venkata Vara Prasad Karri (Visakhapatnam, IN); Tanvi Tayal (White Plains, NY); Shikhar Kwatra (San Jose, CA); Hemant Kumar Sivaswamy (Pune, IN)

Assignee: International Business Machines Corporation

G16H20/60A23L33/30G01N33/02G06N5/04

Abstract

Claims (57)

1 . A computer-implemented method for a personalized virtual assistant interface implemented by an information handling system that includes a processor and a memory, the method comprising:

training an artificial intelligence (AI) system to generate a meal recommendation corresponding to a human subject, the AI system trained with a plurality of eating characteristics corresponding to the human subject, wherein the eating characteristics include one or more eating patterns, a plurality of health data, and a plurality of activity data;

communicating with a plurality of sensors to obtain current food consumption data and current physical activity data for the human subject to include in a feedback loop;

continually training the AI system using the feedback loop to generate the meal recommendation based on the current food consumption data and current physical activity data;

determining a set of caloric needs pertaining to the human subject using current net data for the human subject, the current net data including current consumption data and current energy output data for the human subject provided by one or more sensors;

generating, by the trained AI system, the meal recommendation corresponding to the human subject, wherein the meal recommendation includes a recommended meal time and one or more food recommendations that are based upon the set of caloric needs pertaining to the human subject; and

automatically providing the generated meal recommendation at a time that is based on the recommended meal time at the personalized virtual assistant interface that is voice-enabled and is accessible by the AI system.

2 . The method of claim 1 further comprising:

inputting a set of personal characteristics corresponding to the human subject to the trained AI system;

responsively receiving, from the trained AI system, a crowd-based caloric needs data based on a comparison of the human subject's personal characteristics with a plurality of personal characteristics corresponding to a plurality of other human subjects with sets of corresponding dietary data and the plurality personal characteristics previously learned by the AI system; and

generating the meal recommendation by further incorporating the received crowd-based caloric needs data.

3 . The method of claim 1 further comprising:

receiving a plurality of environmental factors corresponding to a current location of the human subject, wherein at least one of the environmental factors is a set of weather data of the current location, wherein the generating of the meal recommendation is further based on at least one of the received environmental factors.

4 . The method of claim 1 further comprising:

identifying, using one or more sensors, a plurality of food items currently available to a user, wherein the generation of the meal recommendation is further based on the food items currently available to the user; and

updating the plurality of currently available food items based on a usage of the available food items performed by the user.

5 . The method of claim 1 further comprising:

vocally interacting with a user using the voice-enabled virtual assistant, wherein a user provides one or more current individual attributes pertaining to the human subject; and

further training the AI system using the current individual attributes, wherein the user is a caretaker of the human subject.

6 . An information handling system for a personalized virtual assistant interface comprising:

one or more processors;

a memory coupled to at least one of the processors;

a set of computer program instructions stored in the memory and executed by at least one of the processors in order to perform actions comprising:

training an artificial intelligence (AI) system to generate a meal recommendation corresponding to a human subject, the AI system trained with a plurality of eating characteristics corresponding to a human subject, wherein the eating characteristics include one or more eating patterns, a plurality of health data, and a plurality of activity data;

communicating with a plurality of sensors to obtain current food consumption data and current physical activity data for the human subject to include in a feedback loop;

continually training the AI system using the feedback loop to generate the meal recommendation based on the current food consumption data and current physical activity data;

determining a set of caloric needs pertaining to the human subject using current net data for the human subject, the current net data including current consumption data and current energy output data for the human subject provided by one or more sensors;

generating, by the trained AI system, the meal recommendation corresponding to the human subject, wherein the meal recommendation includes a recommended meal time and one or more food recommendations that are based upon the set of caloric needs pertaining to the human subject; and

automatically providing the generated meal recommendation at a time that is based on the recommended meal time at the personalized virtual assistant interface that is voice-enabled and is accessible by the AI system.

7 . The information handling system of claim 6 wherein the actions further comprise:

inputting a set of personal characteristics corresponding to the human subject to the trained AI system;

responsively receiving, from the trained AI system, a crowd-based caloric needs data based on a comparison of the human subject's personal characteristics with a plurality of personal characteristics corresponding to a plurality of other human subjects with sets of corresponding dietary data and the plurality personal characteristics previously learned by the AI system; and

generating the meal recommendation by further incorporating the received crowd-based caloric needs data.

8 . The information handling system of claim 6 wherein the actions further comprise:

receiving a plurality of environmental factors corresponding to a current location of the human subject, wherein at least one of the environmental factors is a set of weather data of the current location, wherein the generating of the meal recommendation is further based on at least one of the received environmental factors.

9 . The information handling system of claim 6 wherein the actions further comprise:

identifying, using one or more sensors, a plurality of food items currently available to a user, wherein the generation of the meal recommendation is further based on the food items currently available to the user; and

updating the plurality of currently available food items based on a usage of the available food items performed by the user.

10 . The information handling system of claim 6 wherein the actions further comprise:

vocally interacting with a user using the voice-enabled virtual assistant, wherein the user provides one or more current individual attributes pertaining to the human subject; and

further training the AI system using the current individual attributes, wherein the user is a caretaker of the human subject.

11 . A computer program product stored in a computer readable storage medium, comprising computer program code that, when executed by an information handling system, performs actions comprising:

training an artificial intelligence (AI) system to generate a meal recommendation corresponding to a human subject, the AI system trained with a plurality of eating characteristics corresponding to a human subject, wherein the eating characteristics include one or more eating patterns, a plurality of health data, and a plurality of activity data;

communicating with a plurality of sensors to obtain current food consumption data and current physical activity data for the human subject to include in a feedback loop;

continually training the AI system using the feedback loop to generate the meal recommendation based on the current food consumption data and current physical activity data;

determining a set of caloric needs pertaining to the human subject using current net data for the human subject, the current net data including current consumption data and current energy output data for the human subject provided by one or more sensors;

generating, by the trained AI system, the meal recommendation corresponding to the human subject, wherein the meal recommendation includes a recommended meal time and one or more food recommendations that are based upon the set of caloric needs pertaining to the human subject; and

automatically providing the generated meal recommendation at a time that is based on the recommended meal time at a personalized virtual assistant interface that is voice-enabled and is accessible by the AI system.

12 . The computer program product of claim 11 wherein the actions further comprise:

inputting a set of personal characteristics corresponding to the human subject to the trained AI system;

responsively receiving, from the trained AI system, a crowd-based caloric needs data based on a comparison of the human subject's personal characteristics with a plurality of personal characteristics corresponding to a plurality of other human subjects with sets of corresponding dietary data and the plurality personal characteristics previously learned by the AI system; and

generating the meal recommendation by further incorporating the received crowd-based caloric needs data.

13 . The computer program product of claim 11 wherein the actions further comprise:

receiving a plurality of environmental factors corresponding to a current location of the human subject, wherein at least one of the environmental factors is a set of weather data of the current location, wherein the generating of the meal recommendation is further based on at least one of the received environmental factors.

14 . The computer program product of claim 11 wherein the actions further comprise:

identifying, using one or more sensors, a plurality of food items currently available to a user, wherein the generation of the meal recommendation is further based on the food items currently available to the user; and

updating the plurality of currently available food items based on a usage of the available food items performed by the user.

Assignments (3)

Continuity (1)

References Cited (31)