Systems and methods for garment size recommendation

1. A computer-implemented method comprising:

determining a first set of features including one or more product attributes of a product selected for a user;

determining a second set of features based on fit ratings for one or more pairs of user identifiers and product identifiers associated with the user;

receiving a query user identifier and a query product identifier associated with the user;

determining a parameterized model associated with the query user identifier and the query product identifier based on the first set of features and the second set of features; and

using the parameterized model to determine a fit score for the query user identifier and the query product identifier.

2. The computer-implemented method of claim 1 , further comprising:

determining a training data set comprising one or more historical data attributes of previously shipped products, wherein each of the historical data attributes is associated with a user identifier and a product identifier used in an electronic transactions platform.

3. The computer-implemented method of claim 2 , wherein determining the first set of features comprises determining the first set of features based on the determined training data set, and

wherein determining the second set of features comprises determining the second set of features based on the determined training data set.

4. The computer-implemented method of claim 2 , further comprising:

training the parameterized model based on the determined training data set, wherein the parameterized model is a generalized linear mixed model (GLMM).

5. The computer-implemented method of claim 2 , wherein determining the second set of features further comprises:

obtaining one or more fit ratings associated with the one or more pairs of user identifiers and product identifiers from the determined training data set;

determining an absence of a fit rating for a particular user identifier and a particular product identifier; and

as a result of determining the absence of the fit rating for the particular pair user identifier and product identifier, determining the fit rating for the particular user identifier and product identifier as negative.

6. The computer-implemented method of claim 1 , wherein determining the parameterized model comprises:

determining (1) one or more fixed effect coefficients, and (2) one or more random effect coefficients, wherein the one or more random effect coefficients are based on variations of a user associated with the query user identifier and a product associated with the query product identifier.

7. The computer-implemented method of claim 1 , further comprising:

comparing the determined fit score for the query user identifier and the query product identifier with another fit score associated with a different user identifier and a different product identifier; and

transmitting the query user identifier and product identifier or the different user identifier and product identifier based on the comparison.

8. A computer system comprising:

a data storage device storing processor-readable instructions; and

a processor configured to execute the instructions to perform a method including:

determining a first set of features including one or more product attributes of a product selected for a user;

determining a second set of features based on fit ratings for one or more pairs of user identifiers and product identifiers associated with the user;

receiving a query user identifier and a query product identifier associated with the user;

determining a parameterized model associated with the query user identifier and the query product identifier based on the first set of features and the second set of features; and

using the parameterized model to determine a fit score for the query user identifier and the query product identifier.

9. The computer system of claim 8 , the method further comprising:

determining a training data set comprising one or more historical data attributes of previously shipped products, wherein each of the historical data attributes is associated with a user identifier and a product identifier used in an electronic transactions platform.

10. The computer system of claim 9 , wherein determining the first set of features comprises determining the first set of features based on the determined training data set, and

wherein determining the second set of features comprises determining the second set of features based on the determined training data set.

11. The computer system of claim 9 , the method further comprising:

training the parameterized model based on the determined training data set, wherein the parameterized model is a generalized linear mixed model (GLMM).

12. The computer system of claim 9 , wherein determining the second set of features further comprises:

obtaining one or more fit ratings associated with the one or more pairs of user identifiers and product identifiers from the determined training data set;

determining an absence of a fit rating for a particular user identifier and a particular product identifier; and

as a result of determining the absence of the fit rating for the particular pair user identifier and product identifier, determining the fit rating for the particular user identifier and product identifier as negative.

13. The computer system of claim 8 , wherein determining the parameterized model comprises:

determining (1) one or more fixed effect coefficients, and (2) one or more random effect coefficients, wherein the one or more random effect coefficients are based on variations of a user associated with the query user identifier and a product associated with the query product identifier.

14. The computer system of claim 8 , the method further comprising:

comparing the determined fit score for the query user identifier and the query product identifier with another fit score associated with a different user identifier and a different product identifier; and

transmitting the query user identifier and product identifier or the different user identifier and product identifier based on the comparison.

15. A non-transitory computer-readable medium containing instructions that, when executed by a processor, cause the processor to perform a method comprising:

determining a first set of features including one or more product attributes of a product selected for a user;

determining a second set of features based on fit ratings for one or more pairs of user identifiers and product identifiers associated with the user;

receiving a query user identifier and a query product identifier associated with the user;

determining a parameterized model associated with the query user identifier and the query product identifier based on the first set of features and the second set of features; and

using the parameterized model to determine a fit score for the query user identifier and the query product identifier.

16. The non-transitory computer-readable medium of claim 15 , the method further comprising:

determining a training data set comprising one or more historical data attributes of previously shipped products, wherein each of the historical data attributes is associated with a user identifier and a product identifier used in an electronic transactions platform.

17. The non-transitory computer-readable medium of claim 16 , wherein determining the first set of features comprises determining the first set of features based on the determined training data set, and

wherein determining the second set of features comprises determining the second set of features based on the determined training data set.

18. The non-transitory computer-readable medium of claim 16 , the method further comprising:

training the parameterized model based on the determined training data set, wherein the parameterized model is a generalized linear mixed model (GLMM),

wherein determining the second set of features further comprises:

obtaining one or more fit ratings associated with the one or more pairs of user identifiers and product identifiers from the determined training data set;

determining an absence of a fit rating for a particular user identifier and a particular product identifier; and

as a result of determining the absence of the fit rating for the particular pair user identifier and product identifier, determining the fit rating for the particular user identifier and product identifier as negative.

19. The non-transitory computer-readable medium of claim 15 , wherein determining the parameterized model comprises:

determining: (1) one or more fixed effect coefficients, and (2) one or more random effect coefficients, wherein the one or more random effect coefficients are based on variations of a user associated with the query user identifier and a product associated with the query product identifier.

20. The non-transitory computer-readable medium of claim 15 , the method further comprising:

comparing the determined fit score for the query user identifier and the query product identifier with another fit score associated with a different user identifier and a different product identifier; and

transmitting the query user identifier and product identifier or the different user identifier and product identifier based on the comparison.