Systems and methods for personalized sizing codes

1 . A method for outputting a size code from a managed database of size codes by a size engine, the method comprising:

receiving, via an extension associated with a user device, a plurality of size nodes for a plurality of users, the plurality of size nodes including at least one of incoming size nodes or outgoing size nodes for one or more items, the one or more items each corresponding to respective item entities, the plurality of size nodes having been generated based on a plurality of interactions monitored by the extension,

wherein the plurality of user interactions are associated with the plurality of users across a plurality of respective extensions;

generating, via a trained machine learning model, a plurality of size profiles based on the plurality of size nodes, wherein:

each of the plurality of size profiles include a unique variation of the plurality of size nodes for the respective item entities,

the plurality of size profiles each correlate a size for each item entity to each other item entity of the respective item entities, and

the trained machine learning model has been trained to predict a size profile based on a plurality of training size nodes and a plurality of training inter-node relationships;

receiving first user size nodes for a first user;

matching the first user to a first size profile of the plurality of size profiles based on the first user size nodes;

receiving a first item corresponding to a first item entity of the item entities;

identifying, at the size engine, the size code for the first item based on the first size profile; and

outputting, via a graphical user interface (GUI) associated with the user device, the size code.

2 . The method of claim 1 , wherein the plurality of size nodes are received at one or more extensions.

3 . The method of claim 2 , wherein the one or more extensions are configured to receive the plurality of size nodes from a respective plurality of point-of-sale portals corresponding to the item entities.

4 . The method of claim 3 , wherein each of the plurality of point-of-sale portals correspond to one or more of a Uniform Resource Locator (URL), a domain, a device application, or a marketplace.

5 . The method of claim 2 , wherein the one or more extensions receive the plurality of size nodes from a respective point-of-sale portal Application Programming Interface (API).

6 . The method of claim 2 , wherein the size code is output via one of the one or more extensions on the user device.

7 . The method of claim 1 , wherein each of the plurality of size nodes includes one or more of an item size, an item fit, an item feedback, an outgoing feedback for a respective item entity, or a re-buy item size.

8 . The method of claim 1 , wherein the plurality of size nodes is received from user inputs.

9 . The method of claim 1 , wherein the outgoing size nodes includes a null outgoing marker.

10 . The method of claim 1 , further comprising determining a correlation score for the first size profile, wherein the first size profile is matched to the first user based on the correlation score being above a minimum correlation score.

11 . The method of claim 1 , wherein at least one of the plurality of size profiles or the matching the first user to the first size profile is implemented using a vector based machine learning model.

12 . A system comprising:

a size engine;

an extension associated with a user device;

a data storage device storing processor-readable instructions; and

a processor operatively connected to the data storage device and the size engine, and configured to execute the processor-readable instructions to perform operations that include:

receiving, via the extension, a plurality of size nodes for a plurality of users, the plurality of size nodes including at least one of incoming size nodes or outgoing size nodes for one or more items, the one or more items each corresponding to respective item entities;

generating, via a trained machine learning model, a plurality of size profiles based on the plurality of size nodes, wherein each of the plurality of size profiles include a unique variation of the plurality of size nodes for the item entities, wherein the plurality of size profiles each correlate a size for each item entity to each other item entity of the item entities, wherein the trained machine learning model has been trained to predict a size profile based on a plurality of training size nodes and a plurality of training inter-node relationships;

receiving a first user size nodes for a first user;

matching the first user to a first size profile of the plurality of size profiles based on the first user size nodes;

receiving a first item corresponding to a first item entity of the item entities;

generating, at the size engine, a size code for the first item based on the first size profile; and

outputting, via a graphical user interface (GUI) associated with the user device, the size code.

13 . The system of claim 12 , wherein:

the plurality of size nodes have been generated based on a plurality of interactions monitored by the extension, and

the plurality of interactions are associated with the plurality of users across a plurality of respective extensions.

14 . The system of claim 13 , wherein the plurality of respective extensions are configured to receive the plurality of size nodes from a respective plurality of point-of-sale portals corresponding to the item entities.

15 . The system of claim 12 , wherein the outgoing size nodes include a null outgoing marker.

16 . A method for outputting a size code from a managed database of size codes by a size engine, the method comprising:

receiving, via an extension associated with a user device, a plurality of size nodes for a plurality of users, the plurality of size nodes having been generated by:

monitoring, via the extension, a plurality of user interactions associated with a plurality of item actions across a plurality of respective extensions, and

generating, via the extension, the plurality of size nodes for the plurality of users based on the plurality of interactions,

the plurality of size nodes including at least one of incoming size nodes or outgoing size nodes for one or more items, the one or more items each corresponding to respective item entities;

receiving, via the extension, a plurality of size profiles based on the plurality of size nodes, wherein:

each of the plurality of size profiles include a unique variation of the plurality of size nodes for the respective item entities, and

the plurality of size profiles each correlate a size for each item entity to each other item entity of the respective item entities;

receiving first user size nodes for a first user;

matching the first user to a first size profile of the plurality of size profiles based on the first user size nodes;

receiving a first item corresponding to a first item entity of the item entities;

identifying, at the size engine, the size code for the first item based on the first size profile; and

outputting, via a graphical user interface (GUI) associated with the user device, the size code.

17 . The method of claim 16 , wherein the item actions include accessing an item page, adding an item to a virtual cart, adding the item to a wish list, or requesting the size code.

18 . The method of claim 16 , wherein each of the plurality of size nodes includes one or more of an item size, an item fit, an item feedback, an outgoing feedback for a respective item entity, or a re-buy item size.

19 . The method of claim 18 , further comprising:

receiving a new size node for a new item entity, the new item entity being different than the item entities;

determining a change coefficient for the new item entity based on the new size node and the plurality of size nodes; and

generating, at the size engine, a plurality of updated size profiles by updating the plurality of size profiles to include the new item entity based on the change coefficient for the new item entity.

20 . The method of claim 16 , further comprising:

generating, via a trained machine learning model, the plurality of size profiles based on the plurality of size nodes, the trained machine learning model having been trained to predict a size profile based on a plurality of training size nodes and a plurality of training inter-node relationships.