Apparatus and method for using a feedback loop to optimize meals
AssigneeKpn Innovations, LLC.
InventorKenneth Neumann
Life of this patent
Estimating the term…
Abstract
The present disclosure is generally directed to an apparatus for using a feedback loop to optimize meals, may include at least a processor; and a memory communicatively connected to the processor, the memory containing instructions configuring the at least a processor to retrieve nutrition data from a database. The processor may be configured to generate an optimization score, wherein generating the optimization score may include training an optimization machine-learning model, wherein the optimization machine-learning model is trained with optimization training data, inputting a nutrient quantity to the optimization machine-learning model to output a target nutrient score, and generating an optimization score as a function of the nutrition data and the target nutrient score.
Claims20
An apparatus for using a feedback loop to optimize meals, the apparatus comprising:
- at least a processor;
- and
- a memory communicatively connected to the processor, the memory containing instructions configuring the at least a processor to:
- retrieve nutrition data for a plurality of users sharing a phenotype from a database;
- generate a plurality of nutrient scores using a nutrient classifier, wherein the plurality of nutrient scores relate to an impact of one or more nutrients on one or more phenotypes, wherein generating the plurality of nutrient scores comprises:
- training the nutrient classifier using nutrient training data, wherein the nutrient training data comprises historical nutrient data correlated to categories of nutrients;
- generating the plurality of nutrient scores as a function of the nutrient classifier;
- generate an optimization score, wherein generating the optimization score comprises:
- training an optimization machine-learning model, wherein the optimization machine-learning model is trained using optimization training data;
- inputting the nutrition data to the optimization machine-learning model to output a target nutrient score, wherein the target nutrient score comprises a nutrient score for a particular phenotype associated with a user from the plurality of nutrient scores;
- and
- generating the optimization score as a function of the nutrition data and the target nutrient score;
- and
- modify the nutrition data as a function of the optimization score, wherein modifying the nutrition data comprises reassigning at least one user to a different phenotype.
A method for using a feedback loop to optimize meals, the method comprising:
- retrieving, by a processor, nutrition data for a plurality of users sharing a phenotype from a database;
- generating, by the processor, a plurality of nutrient scores using a nutrient classifier, wherein the plurality of nutrient scores relate to an impact of one or more nutrients on one of more phenotypes, wherein generating the nutrient score comprises:
- training the nutrient classifier using nutrient training data, wherein the nutrient training data comprises historical nutrient data correlated to categories of nutrients;
- generating the plurality of nutrient scores as a function of the nutrient classifier;
- generating, by the processor, an optimization score, wherein generating the optimization score comprises:
- training an optimization machine-learning model, wherein the optimization machine-learning model is trained with optimization training data;
- inputting a nutrient quantity to the optimization machine-learning model to output a target nutrient score, wherein the target nutrient score comprises a nutrient score for a particular phenotype associated with a user from the plurality of nutrient scores;
- and
- generating the optimization score as a function of the nutrition data and the target nutrient score;
- and
- modifying, by the processor, the nutrition data as a function of the optimization score, wherein modifying the nutrition data comprises reassigning at least one user to a different phenotype.
Dependencies are inferred from the claim text.