Systems, methods and computer program products for cross-marketing related products and services based on machine learning algorithms involving field identifier level adjacencies
Embodiments disclosed herein can leverage dynamic data, fields and implied information from online media (e.g., web sites, mobile apps and other forms of online media) to create a universal variable map and automatically generate deep-linked ads utilizing the universal variable map, linking any arbitrary online media to any other arbitrary online media. The automatically generated deep-linked ads can be inserted into any ad server and served out from there in real-time. A deep-linked ad thus presented to a user can take the user from a starting page on one site or application where the ad is displayed directly to a resulting page such as a checkout page on another site or mobile application, with pertinent information already pre-populated, and action already initiated for the user.
1. A method, comprising:
algorithmically setting and refining adjacency proximities of field identifiers or their values across similar or different verticals of products or services on the Internet, the field identifiers mapped through a universal variable data database, the universal variable data database representing a universal variable map of searchable classes and attributes of products or services on the Internet, the algorithmically setting and refining performed by a computer having a processor and a non-transitory computer-readable medium, the universal variable data database maintained by the computer;
storing the adjacency proximities thus determined by the computer in an adjacency matrix;
generating, by the computer using the universal variable data database, the adjacency matrix, and any specified business rules, a cross-marketing deep-linked ad that links field identifiers of different products or services;
placing the cross-marketing deep-linked ad in a first site, application, or online service delivered to a user device, the first site, application, or online service associated with a first product or service, the cross-marketing deep-linked ad presenting a second product or service;
in response to a user interacting with the cross-marketing deep-linked ad presented on the user device, passing dynamic data in field identifiers of the first site, application, or online service to a second site, application, or online service associated with the second product or service such that the second site, application, or online service can present information determined or pre-populated based on the dynamic data from the first site, application, or online service without necessitating any pre-existing relationship between the first and second sites, applications, or online services; and
wherein the first and second sites, applications or online services belong to the same or different verticals.
2. The method according to claim 1 , wherein the computer performs the algorithmically setting and refining based on consumer click-through rates of cross-marketed offerings.
3. The method according to claim 1 , wherein the algorithmically setting and refining further comprises adjusting the adjacency proximities of field identifiers or their values relative to a user engagement and action rate, wherein the computer learns, using machine learning, the user engagement and action rate from data collected by the computer over time.
4. The method according to claim 1 , wherein the adjacency matrix is initially seeded based on a category analysis of the products or services on the Internet or by setting adjacencies based on initial user interactions of any or all possible permutations of adjacency proximities.
5. The method according to claim 1 , wherein the adjacency matrix is initially seeded based on a number of matches between field identifiers across any pair of product or service categories.
6. A system, comprising:
a processor;
a non-transitory computer-readable medium; and
stored instructions translatable by the processor to perform:
algorithmically setting and refining adjacency proximities of field identifiers or their values across similar or different verticals of products or services on the internet, the field identifiers mapped through a universal variable data database, the universal variable data database representing a universal variable map of searchable classes and attributes of products or services on the Internet, the algorithmically setting and refining performed by a computer having a processor and a non-transitory computer-readable medium, the universal variable data database maintained by the system;
storing the adjacency proximities thus determined in an adjacency matrix;
generating, using the universal variable data database, the adjacency matrix, and any specified business rules, a cross-marketing deep-linked ad that links field identifiers of different products or services;
placing the cross-marketing deep-linked ad in a first site, application, or online service delivered to a user device, the first site, application, or online service associated with a first product or service, the cross-marketing deep-linked ad presenting a second product or service;
in response to a user interacting with the cross-marketing deep-linked ad presented on the user device, passing dynamic data in field identifiers of the first site, application, or online service to a second site, application, or online service associated with the second product or service such that the second site, application, or online service can present information determined or pre-populated based on the dynamic data from the first site, application, or online service without necessitating any pre-existing relationship between the first and second sites, applications, or online services; and
wherein the first and second sites, applications or online services belong to the same or different verticals.
7. The system of claim 6 , wherein the algorithmically setting and refining is performed based on consumer click-through rates of cross-marketed offerings.
8. The system of claim 6 , wherein the algorithmically setting and refining further comprises adjusting the adjacency proximities of field identifiers or their values relative to a user engagement and action rate, wherein the system learns, using machine learning, the user engagement and action rate from data collected by the system over time.
9. The system of claim 6 , wherein the adjacency matrix is initially seeded based on a category analysis of the products or services on the Internet or by setting adjacencies based on initial user interactions of any or all possible permutations of adjacency proximities.
10. The system of claim 6 , wherein the adjacency matrix is initially seeded based on a number of matches between field identifiers across any pair of product or service categories.
11. A computer program product comprising a non-transitory computer-readable medium storing instructions translatable by a processor of a computer system to perform:
algorithmically setting and refining adjacency proximities of field identifiers or their values across similar or different verticals of products or services on the internet, the field identifiers mapped through a universal variable data database, the universal variable data database representing a universal variable map of searchable classes and attributes of products or services on the Internet, the algorithmically setting and refining performed by a computer having a processor and a non-transitory computer-readable medium, the universal variable data database maintained by the computer system;
storing the adjacency proximities thus determined in an adjacency matrix;
generating, using the universal variable data database, the adjacency matrix, and any specified business rules, a cross-marketing deep-linked ad that links field identifiers of different products or services;
placing the cross-marketing deep-linked ad in a first site, application, or online service delivered to a user device, the first site, application, or online service associated with a first product or service, the cross-marketing deep-linked ad presenting a second product or service;
in response to a user interacting with the cross-marketing deep-linked ad presented on the user device, passing dynamic data in field identifiers of the first site, application, or online service to a second site, application, or online service associated with the second product or service such that the second site, application, or online service can present information determined or pre-populated based on the dynamic data from the first site, application, or online service without necessitating any pre-existing relationship between the first and second sites, applications, or online services; and
wherein the first and second sites, applications or online services belong to the same or different verticals.
12. The computer program product of claim 11 , wherein the algorithmically setting and refining is performed based on consumer click-through rates of cross-marketed offerings.
13. The computer program product of claim 11 , wherein the adjacency matrix is initially seeded based on a category analysis of the products or services on the Internet or by setting adjacencies based on initial user interactions of any or all possible permutations of adjacency proximities.
14. The computer program product of claim 11 , wherein the adjacency matrix is initially seeded based on a number of matches between field identifiers across any pair of product or service categories.