Crowd-based personalized recommendations of food using measurements of affective response

• US 10,387,898 B2
• Filed: 09/11/2017
• Issued: 08/20/2019
• Est. Priority Date: 08/21/2014
• Status: Active Grant

• Current Assignee: Affectomatics ltd.
• Original Assignee: Affectomatics ltd.
• Inventors: Frank, Ari M, Thieberger, Gil
• Primary Examiner(s): Moser, Bruce M
• Application Number:
  US15/701,101
• Publication Number:
  US 20180025368A1
• Time in Patent Office: 708 Days
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/337 Profile generation, learning or modification
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 10/101 Collaborative creation, e.g. joint development of products or services
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0282 Rating or review of business operators or products
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: G06Q 50/12 Hotels or restaurants

Abstract

Some aspects of this disclosure involve ranking types of food based on measurements of affective response of users who consumed the types of food. In one embodiment, a system that ranks types of food includes sensors that take measurements of affective response of users and a computer. The computer receives indications of times at which the users consumed the types of food and selects for each type of food, based on the indications, measurements of at least five of the users taken up to four hours after consuming the type of food. The computer ranks the types of food based on the selected measurements. Optionally, each of the types of food is a different item on a menu comprising food items.

Claims

1. A system configured to make personalized recommendations of types of food, comprising:

sensors configured to take measurements of affective response of users; and

a computer configured to;

receive indications of times at which the users consumed the types of food;

select for each type of food, based on the indications, measurements of at least five of the users taken up to four hours after consuming the type of food;

receive profiles of the users and first and second profiles of first and second users, respectively;

generate a first output indicative of similarities between the first profile and the profiles of the users;

calculate a first ranking of the types of food based on the measurements and the first output;

recommend to the first user, based on the first ranking, a first type of food over a second type of food;

generate a second output indicative of similarities between the second profile and the profiles of the users;

calculate a second ranking of the types of food based on the measurements and the second output; and

recommend to the second user, based on the second ranking, the second type of food over the first type of food.

2. The system of claim 1, wherein on average, the at least five of the users who consumed the first type of food had a more positive affective response than the at least five of the users who consumed the second type of food.

3. The system of claim 1, wherein each of the types of food is a different item on a menu comprising food items.

4. The system of claim 1, wherein the computer is further configured to:

(i) receive information, obtained from a financial account of a user from among the users, indicating that the user purchased the type of food from among the types of food; and

(ii) select, based on the information, which measurements of affective response of the user to utilize for ranking the types of food.

5. The system of claim 1, wherein the computer is further configured to:

(i) receive information, obtained from a social media account of a user from among the at least five of the users who consumed a certain type of food, indicating when the user consumed the certain type of food; and

(ii) select, based on the information, which measurements of affective response of the user to utilize for ranking the types of food.

6. The system of claim 1, wherein a profile of a user comprises indications of one or more of the following:

(i) an indication of a preferred dietary choice comprising one or more of the following;

vegetarian food, vegan food, and low-carb food;

(ii) an indication of a sensitivity to a certain food ingredient;

(iii) a food allergy;

(iv) a preference regarding a certain brand;

(v) a preference towards a certain restaurant;

(vi) and an extent of previous consumption of one or more types of food.

7. The system of claim 1, wherein the computer is configured to generate the first output by calculating values indicative of a similarity between each of the profiles of the users and the first profile, and utilizing the values to calculate weights for the measurements of the users;

wherein a weight for a measurement of a user is proportional to an extent of a similarity between the profile of the user and the profile of the first user, such that a weight calculated for a measurement of a user whose profile is more similar to the profile of the first user is higher than a weight calculated for a measurement of a user whose profile is less similar to the profile of the first user;

wherein the first output is indicative of the weights for the measurements of the users; and

wherein the computer is configured to;

calculate for each type of food a score based on the measurements of the at least five of the users who consumed the type of food and the weights calculated for said measurements, and to calculate the first ranking of the types of food based on magnitudes of the scores for the types of food;

wherein the score calculated for the first type of food is higher than the score calculated for the second type of food.

8. The system of claim 1, wherein the computer is configured to generate the first output by calculating values indicative of a similarity between each of the profiles and the first profile, and utilizing the values to calculate weights for the measurements of the users;

wherein the weight for a measurement of a user is proportional to an extent of a similarity between the profile of the user and the profile of the first user, such that a weight calculated for a measurement of a user whose profile is more similar to the profile of the first user is higher than a weight calculated for a measurement of a user whose profile is less similar to the profile of the first user;

the first output is indicative of the weights for the measurements of the users;

the computer is further configured to generate a plurality of preference rankings for the types of food;

each preference ranking is determined based on a subset of the measurements, has a corresponding weight determined based on the weights of the measurements comprised in the subset, and comprises a ranking of at least two of the types of food, such that one of the at least two types of food is ranked ahead of another type of food from among the at least two types of food; and

wherein the computer is further configured to calculate the first ranking of the types of food based on the plurality of the preference rankings utilizing a method that satisfies the Condorcet criterion.

9. The system of claim 1, wherein the computer is configured to generate the first output by clustering users into clusters based on similarities between the profiles of the users, with each cluster comprising a single user or multiple users with similar profiles, and selecting, based on the first profile, a subset comprising at most half of the clusters;

wherein, on average, the first profile is more similar to a profile of a user who is a member of a cluster in the subset, than it is to a profile of a user, from among the users, who is not a member of any of the clusters in the subset;

the computer is further configured to select at least eight users from among the users belonging to clusters in the subset;

wherein the first output is indicative of the at least eight users; and

wherein the computer is configured to calculate the first ranking of the types of food based on the measurements of the at least eight users.

10. A method for making personalized recommendations of types of food, comprising:

taking measurements of affective response of users with sensors;

receiving indications of times at which the users consumed the types of food;

selecting for each type of food, based on the indications, measurements of at least five of the users taken up to four hours after consuming the type of food;

receiving profiles of the users and first and second profiles of first and second users, respectively;

generating a first output indicative of similarities between the first profile and the profiles of the users;

calculating a first ranking of the types of food based on the measurements and the first output;

recommending to the first user, based on the first ranking, a first type of food over a second type of food;

generating a second output indicative of similarities between the second profile and the profiles of the users;

calculating a second ranking of the types of food based on the measurements and the second output; and

recommending to the second user, based on the second ranking, the second type of food over the first type of food.

11. The method of claim 10, further comprising:

(i) receiving information, obtained from a financial account of a user from among the users, indicating that the user purchased the type of food from among the types of food;

and (ii) selecting, based on the information, which measurements of affective response of the user to utilize for ranking the types of food.

12. The method of claim 10, further comprising:

(i) receiving information, obtained from a social media account of a user from among the at least five of the users who consumed a certain type of food, indicating when the user consumed the certain type of food; and

(ii) selecting, based on the information, which measurements of affective response of the user to utilize for ranking the types of food.

13. The method of claim 10, wherein a profile of a user comprises indications of one or more of the following:

(i) an indication of a preferred dietary choice comprising one or more of the following;

vegetarian food, vegan food, and low-carb food;

(ii) an indication of a sensitivity to a certain food ingredient;

(iii) a food allergy;

(iv) a preference regarding a certain brand;

(v) a preference towards a certain restaurant;

(vi) and an extent of previous consumption of one or more types of food.

14. The method of claim 10, further comprising:

generating the first output by calculating values indicative of a similarity between each of the profiles of the users and the first profile;

utilizing the values to calculate weights for the measurements of the users;

wherein a weight for a measurement of a user is proportional to an extent of a similarity between the profile of the user and the profile of the first user, such that a weight calculated for a measurement of a user whose profile is more similar to the profile of the first user is higher than a weight calculated for a measurement of a user whose profile is less similar to the profile of the first user;

wherein the first output is indicative of the weights for the measurements of the users;

calculating for each type of food a score based on the measurements of the at least five of the users who consumed the type of food and the weights calculated for said measurements; and

generating the first ranking of the types of food based on magnitudes of the scores for the types of food;

wherein the score calculated for the first type of food is higher than the score calculated for the second type of food.

15. The method of claim 10, further comprising:

generating the first output by calculating values indicative of a similarity between each of the profiles and the first profile;

utilizing the values to calculate weights for the measurements of the users;

wherein the weight for a measurement of a user is proportional to an extent of a similarity between the profile of the user and the profile of the first user, such that a weight calculated for a measurement of a user whose profile is more similar to the profile of the first user is higher than a weight calculated for a measurement of a user whose profile is less similar to the profile of the first user;

the first output is indicative of the weights for the measurements of the users;

generating a plurality of preference rankings for the types of food;

each preference ranking is determined based on a subset of the measurements, has a corresponding weight determined based on the weights of the measurements comprised in the subset, and comprises a ranking of at least two of the types of food, such that one of the at least two types of food is ranked ahead of another type of food from among the at least two types of food; and

calculating the first ranking of the types of food based on the plurality of the preference rankings utilizing a method that satisfies the Condorcet criterion.

16. A non-transitory computer-readable medium having instructions stored thereon that, in response to execution by a system including a processor and memory, causes the system to perform operations comprising:

receiving measurements of affective response of users taken with sensors;

receiving indications of times at which the users consumed types of food;

selecting for each type of food, based on the indications, measurements of at least five of the users taken up to four hours after consuming the type of food;

receiving profiles of the users and first and second profiles of first and second users, respectively;

generating a first output indicative of similarities between the first profile and the profiles of the users;

calculating a first ranking of the types of food based on the measurements and the first output;

recommending to the first user, based on the first ranking, a first type of food over a second type of food;

generating a second output indicative of similarities between the second profile and the profiles of the users;

calculating a second ranking of the types of food based on the measurements and the second output; and

recommending to the second user, based on the second ranking, the second type of food over the first type of food.

17. The non-transitory computer-readable medium of claim 16, further comprising additional instructions that, in response to execution, cause the system to perform operations comprising:

generating the first output by clustering users into clusters based on similarities between the profiles of the users, with each cluster comprising a single user or multiple users with similar profiles, and selecting, based on the first profile, a subset comprising at most half of the clusters;

wherein, on average, the first profile is more similar to a profile of a user who is a member of a cluster in the subset, than it is to a profile of a user, from among the users, who is not a member of any of the clusters in the subset;

selecting at least eight users from among the users belonging to clusters in the subset;

wherein the first output is indicative of the at least eight users; and

calculating the first ranking of the types of food based on the measurements of the at least eight users.

18. The non-transitory computer-readable medium of claim 16, further comprising additional instructions that, in response to execution, cause the system to perform operations comprising:

generating the first output by calculating values indicative of a similarity between each of the profiles of the users and the first profile;

utilizing the values to calculate weights for the measurements of the users;

wherein a weight for a measurement of a user is proportional to an extent of a similarity between the profile of the user and the profile of the first user, such that a weight calculated for a measurement of a user whose profile is more similar to the profile of the first user is higher than a weight calculated for a measurement of a user whose profile is less similar to the profile of the first user;

wherein the first output is indicative of the weights for the measurements of the users;

calculating for each type of food a score based on the measurements of the at least five of the users who consumed the type of food and the weights calculated for said measurements; and

generating the first ranking of the types of food based on magnitudes of the scores for the types of food;

wherein the score calculated for the first type of food is higher than the score calculated for the second type of food.

19. The non-transitory computer-readable medium of claim 16, further comprising additional instructions that, in response to execution, cause the system to perform operations comprising:

generating the first output by calculating values indicative of a similarity between each of the profiles and the first profile;

utilizing the values to calculate weights for the measurements of the users;

wherein the weight for a measurement of a user is proportional to an extent of a similarity between the profile of the user and the profile of the first user, such that a weight calculated for a measurement of a user whose profile is more similar to the profile of the first user is higher than a weight calculated for a measurement of a user whose profile is less similar to the profile of the first user;

the first output is indicative of the weights for the measurements of the users;

generating a plurality of preference rankings for the types of food;

each preference ranking is determined based on a subset of the measurements, has a corresponding weight determined based on the weights of the measurements comprised in the subset, and comprises a ranking of at least two of the types of food, such that one of the at least two types of food is ranked ahead of another type of food from among the at least two types of food; and

calculating the first ranking of the types of food based on the plurality of the preference rankings utilizing a method that satisfies the Condorcet criterion.

20. The non-transitory computer-readable medium of claim 16, further comprising additional instructions that, in response to execution, cause the system to perform operations comprising:

(i) receiving information, obtained from a financial account of a user from among the users, indicating that the user purchased the type of food from among the types of food; and

(ii) selecting, based on the information, which measurements of affective response of the user to utilize for ranking the types of food.