Machine learning for nutrient quantity estimation in score-based diets and methods of use thereof
Systems and methods of the present disclosure use one or more processor(s) to receive a consumable preference and a daily score intake value associated with a user and to obtain a content data regarding consumable item including an amount of a first nutrient found in the consumable item. The processor(s) utilizes, in real-time, a nutrient prediction machine learning model to ingest the content data regarding the consumable item and predict an amount of a second nutrient in the consumable item based on the content data and a decision tree library of thousand nutrient decision trees. The processor(s) determines zero-scored consumable item based on the daily score intake value, the amount of the first nutrient, the amount of the second nutrient, and the consumable preference. The processor(s) instructs a computing device to utilize a graphical user interface element to identify the zero-scored consumable item on a screen of the computing device.
1. A computer-based method comprising:
receiving, by the processor, a content data regarding at least one consumable item;
wherein the at least one consumable item is at least one food item, at least one beverage item, or a combination thereof;
wherein the content data comprises a first content data item identifying at least one first amount of at least one first nutrient found in the at least one consumable item;
applying, by the at least one processor, each decision tree in a subset of decision trees to the content data to predict at least one decision tree-specific second amount of at least one second nutrient in the at least one consumable item based at least in part on the first content data item identifying the at least one first amount of the at least one first nutrient found in the at least one consumable item;
wherein each decision tree of the subset of decision trees comprises a plurality of branches and a plurality of leaves;
wherein the subset of decision trees comprises at least one thousand nutrient decision trees; and
wherein the decision tree model is trained to form each branch of the plurality of branches based at least in part on learned threshold values of nutrient quantities;
aggregating, by the at least one processor, the at least one decision tree-specific second amount of each decision tree to predict at least one second amount of the at least one second nutrient;
determining, by the processor, the at least one consumable item as at least one zero-scored consumable item, or a score value for the at least one consumable item, based, at least in part, on:
the at least one first amount of the at least one first nutrient,
the at least one second amount of the at least one second nutrient; and
instructing, by the processor, the computing device associated with the user to utilize at least one graphical user interface element on a screen of the computing device to display at least one of:
the at least one consumable item as the at least one zero-scored consumable item, or
the score value for the at least one consumable item.
2. The method of claim 1 , further comprising:
receiving, by the at least one processor, at least one known second amount representative of at least one actual amount of the at least one second nutrient in the at least one consumable item;
determining, by the at least one processor, at least one second nutrient prediction error based at least in part on a different between the at least one second amount and the at least one known second amount; and
updating, by the at least one processor, the learned threshold values of nutrient quantities of each branch of the plurality of branches of each decision tree in the subset of decision trees based at least in part on the at least one second nutrient prediction error.
3. The method of claim 2 , further comprising pruning, by the at least one processor, the subset of decision trees based at least in part on the at least one second nutrient prediction error.
4. The method of claim 1 , further comprising:
selecting, by the at least one processor, the subset of decision trees from the decision tree library based at least in part on model parameters;
receiving, by the at least one processor, at least one known second amount representative of at least one actual amount of the at least one second nutrient in the at least one consumable item;
determining, by the at least one processor, at least one second nutrient prediction error based at least in part on a different between the at least one second amount and the at least one known second amount; and
updating, by the at least one processor, the model parameters based at least in part on the at least one second nutrient prediction error.
5. The method of claim 1 , wherein the nutrient prediction machine learning model comprises a random forest.
6. The method of claim 1 , further comprising:
instructing, by the processor, the computing device associated with the user to generate a consumable preference questionnaire;
receiving, by the processor, from the computing device associated with the user, answers to the consumable preference questionnaire; and
generating, by the processor, the daily score intake value associated with the user based, at least in part on, on the answers to the consumable preference questionnaire.
7. The method of claim 1 , further comprising utilizing, by the at least one processor, at least one consumable item intake score based at least in part on the at least one first nutrient amount and the at least on second nutrient amount.
8. The method of claim 7 , further comprising:
receiving, by the at least one processor, at least one consumption indication indicative of the user having consumed the at least one consumable item; and
updating, by the at least one processor, the daily score intake value by deducting the at least one consumable item intake score in response to the at least one consumption indication.
9. A computer-based system comprising:
at least one processor in communication with at least one non-transitory computer readable medium having software instructions stored thereon, wherein, upon execution of the software instructions, the at least one processor is configured to:
receive a content data regarding at least one consumable item;
wherein the at least one consumable item is at least one food item, at least one beverage item, or a combination thereof;
wherein the content data comprises a first content data item identifying at least one first amount of at least one first nutrient found in the at least one consumable item;
apply each decision tree in a subset of decision trees to the content data to predict at least one decision tree-specific second amount of at least one second nutrient in the at least one consumable item based at least in part on the first content data item identifying the at least one first amount of the at least one first nutrient found in the at least one consumable item;
wherein each decision tree of the subset of decision trees comprises a plurality of branches and a plurality of leaves;
wherein the subset of decision trees comprises at least one thousand nutrient decision trees; and
wherein the decision tree model is trained to form each branch of the plurality of branches based at least in part on learned threshold values of nutrient quantities;
aggregate the at least one decision tree-specific second amount of each decision tree to predict at least one second amount of the at least one second nutrient;
determine the at least one consumable item as at least one zero-scored consumable item, or a score value for the at least one consumable item, based, at least in part, on:
the at least one first amount of the at least one first nutrient,
the at least one second amount of the at least one second nutrient; and
instruct the computing device associated with the user to utilize at least one graphical user interface element on a screen of the computing device to display at least one of:
the at least one consumable item as the at least one zero-scored consumable item, or
the score value for the at least one consumable item.
10. The system of claim 9 , wherein, upon execution of the software instructions, the at least one processor is further configured to:
receive at least one known second amount representative of at least one actual amount of the at least one second nutrient in the at least one consumable item;
determine at least one second nutrient prediction error based at least in part on a different between the at least one second amount and the at least one known second amount; and
update the learned threshold values of nutrient quantities of each branch of the plurality of branches of each decision tree in the subset of decision trees based at least in part on the at least one second nutrient prediction error.
11. The system of claim 10 , wherein, upon execution of the software instructions, the at least one processor is further configured to prune the subset of decision trees based at least in part on the at least one second nutrient prediction error.
12. The system of claim 9 , wherein, upon execution of the software instructions, the at least one processor is further configured to:
select the subset of decision trees from the decision tree library based at least in part on model parameters;
receive at least one known second amount representative of at least one actual amount of the at least one second nutrient in the at least one consumable item;
determine at least one second nutrient prediction error based at least in part on a different between the at least one second amount and the at least one known second amount; and
update the model parameters based at least in part on the at least one second nutrient prediction error.
13. The system of claim 9 , wherein the nutrient prediction machine learning model comprises a random forest.
14. The system of claim 9 , wherein, upon execution of the software instructions, the at least one processor is further configured to:
instruct the computing device associated with the user to generate a consumable preference questionnaire;
receive from the computing device associated with the user, answers to the consumable preference questionnaire; and
generate the daily score intake value associated with the user based, at least in part on, on the answers to the consumable preference questionnaire.
15. The system of claim 9 , wherein, upon execution of the software instructions, the at least one processor is further configured to utilize at least one consumable item intake score based at least in part on the at least one first nutrient amount and the at least on second nutrient amount.
16. The system of claim 15 , wherein, upon execution of the software instructions, the at least one processor is further configured to:
receive at least one consumption indication indicative of the user having consumed the at least one consumable item; and
update the daily score intake value by deducting the at least one consumable item intake score in response to the at least one consumption indication.