Predictive recommendation system using absolute relevance
In general, embodiments of the present invention provide systems, methods and computer readable media for ranking promotions selected for recommendation to consumers based on predictions of promotion performance and consumer behavior. In embodiments, a set of promotions to be recommended to a consumer can be sorted and/or ranked according to respective relevance scores representing a probability that the consumer's behavior in response to the promotion will match a ranking target. In embodiments, calculating scores is based on a relevance model (a predictive function) derived from one or more contextual data sources representing attributes of promotions and consumer behavior. In embodiments, an absolute relevance score represents an absolute prediction of a ranking target variable. In embodiments, absolute relevance may be used to determine personalized local merchant discovery frontiers; featured result set thresholding for impressions; and/or promotion notification triggers. In embodiments, predictive models based on gross revenue may be optimized using promotion category-dependent price boosting.
1. A computer-implemented method, comprising:
receiving, from a computing device associated with a consumer, input data representing a request, the input data comprising a ranking target variable and a consumer location associated with multi-tier promotion categories;
receiving, from a user profile repository, user data describing attributes of the consumer;
receiving, from a promotion repository, promotion data associated with the multi-tier promotion categories, the promotion data describing attributes of each available promotion of a set of available promotions associated with the multi-tier promotion categories;
generating an initial absolute relevance score for each available promotion of the set of available promotions using the user data and the promotion data, wherein the initial absolute relevance score represents an estimated absolute relevance probability associated with the ranking target variable;
generating a subset of available promotions by selecting available promotions each having a respective initial absolute relevance score above an absolute relevance score threshold;
determining an optimal promotion category tier-level for the consumer based on the consumer location and a total count of available promotions in the subset of available promotions;
generating an adjusted subset of available promotions based on the subset of available promotions and the optimal promotion category tier-level; and
transmitting the adjusted subset of available promotions to the computing device associated with the consumer.
2. The computer-implemented method of claim 1 , wherein the ranking target variable is associated with a probability that the consumer will purchase a particular promotion using the computing device or a probability that the consumer will click on an impression of a particular promotion using the computing device.
3. The computer-implemented method of claim 1 , further comprising:
ranking the adjusted subset of available promotions based on their respective initial absolute relevance scores.
4. The computer-implemented method of claim 3 , further comprising:
generating at least one personalized notification trigger for the consumer using the ranked adjusted subset of available promotions, the promotion data, and the user data.
5. The computer-implemented method of claim 1 , further comprising:
receiving initialization parameters including a target number of promotions to be selected, a minimum estimated absolute relevance probability, and a promotion category tier-level associated with the consumer location;
receiving a first subset of available promotions each associated with the promotion category tier-level of the multi-tier promotion categories;
calculating a first count of the first subset of available promotions that are associated with an absolute relevance score greater than or equal to the minimum estimated absolute relevance probability; and
in circumstances where the first count is greater than or equal to a scores count threshold, and in circumstances where the first count is less than or equal to the target number of promotions,
selecting the first subset of available promotions to be ranked.
6. The computer-implemented method of claim 5 , further comprising:
in circumstances where the first count is less than the scores count threshold,
increasing the promotion category tier-level to include a promotion subcategory of the multi-tier promotion categories;
receiving a second subset of available promotions each associated with the increased promotion category tier-level of the multi-tier promotion categories; and
calculating a second count of the second subset of available promotions that are associated with an absolute relevance score that is greater than or equal to the minimum estimated absolute relevance probability.
7. The computer-implemented method of claim 5 , further comprising:
in circumstances where the first count is greater than the target number of promotions,
decreasing the promotion category tier-level to exclude a promotion subcategory of the multi-tier promotion categories;
receiving a third subset of available promotions each associated with the decreased promotion category tier-level of the multi-tier promotion categories; and
calculating a third count of the third subset of available promotions that are associated with an absolute relevance score that is greater than or equal to the minimum estimated absolute relevance probability.
8. The computer-implemented method of claim 1 , wherein determining the optimal promotion category tier-level for the consumer includes determining a personalized promotion category multi-tier search for the consumer.
9. A computer program product, encoded on a computer-readable medium, operable to cause data processing apparatus to perform operations comprising:
receiving, from a computing device associated with a consumer, input data representing a request, the input data comprising a ranking target variable and a consumer location associated with multi-tier promotion categories;
receiving, from a user profile repository, user data describing attributes of the consumer;
receiving, from a promotion repository, promotion data associated with the multi-tier promotion categories, the promotion data describing attributes of each available promotion of a set of available promotions associated with the multi-tier promotion categories;
generating an initial absolute relevance score for each available promotion of the set of available promotions using the user data and the promotion data, wherein the initial absolute relevance score represents an estimated absolute relevance probability associated with the ranking target variable;
generating a subset of available promotions by selecting available promotions each having a respective initial absolute relevance score above an absolute relevance score threshold;
determining an optimal promotion category tier-level for the consumer based on the consumer and a total count of available promotions in the subset of available promotions;
generating an adjusted subset of available promotions based on the subset of available promotions and the optimal promotion category tier-level; and
transmitting the adjusted subset of available promotions to the computing device associated with the consumer.
10. The computer program product of claim 9 , wherein the ranking target variable is associated with a probability that the consumer will purchase a particular promotion using the computing device or a probability that the consumer will click on an impression of a particular promotion using the computing device.
11. The computer program product of claim 9 , further comprising:
ranking the adjusted subset of available promotions based on their respective initial absolute relevance scores.
12. The computer program product of claim 11 , further comprising:
generating at least one personalized notification trigger for the consumer using the ranked adjusted subset of available promotions, the promotion data, and the user data.
13. The computer program product of claim 9 , further comprising:
receiving initialization parameters including a target number of promotions to be selected, a minimum estimated absolute relevance probability, and a promotion category tier- level associated with the consumer location;
receiving a first subset of available promotions each associated with the promotion category tier-level of the multi-tier promotion categories;
calculating a first count of the first subset of available promotions that are associated with an absolute relevance score greater than or equal to the minimum estimated absolute relevance probability; and
in circumstances where the first count is greater than or equal to a scores count threshold, and in circumstances where the first count is less than or equal to the target number of promotions,
selecting the first subset of available promotions to be ranked.
14. The computer program product of claim 13 , further comprising:
in circumstances where the first count is less than the scores count threshold,
increasing the promotion category tier-level to include a promotion subcategory of the multi-tier promotion categories;
receiving a second subset of available promotions each associated with the increased promotion category tier-level of the multi-tier promotion categories; and
calculating a second count of the second subset of available promotions that are associated with an absolute relevance score that is greater than or equal to the minimum estimated absolute relevance probability.
15. The computer program product of claim 13 , further comprising:
in circumstances where the first count is greater than the target number of promotions,
decreasing the promotion category tier-level to exclude a promotion subcategory of the multi-tier promotion categories;
receiving a third subset of available promotions each associated with the decreased promotion category tier-level of the multi-tier promotion categories; and
calculating a third count of the third subset of available promotions that are associated with an absolute relevance score that is greater than or equal to the minimum estimated absolute relevance probability.
16. The computer program product of claim 9 , wherein determining the optimal promotion category tier-level for the consumer includes determining a personalized promotion category multi-tier search for the consumer.
17. A system comprising one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations, comprising:
receiving, from a computing device associated with a consumer, input data representing a request, the input data comprising a ranking target variable and a consumer location associated with multi-tier promotion categories;
receiving, from a user profile repository, user data describing attributes of the consumer;
receiving, from a promotion repository, promotion data associated with the multi-tier promotion categories, the promotion data describing attributes of each available promotion of a set of available promotions associated with the multi-tier promotion categories;
generating an initial absolute relevance score for each available promotion of the set of available promotions using the user data and the promotion data, wherein the initial absolute relevance score represents an estimated absolute relevance probability associated with the ranking target variable;
generating a subset of available promotions by selecting available promotions each having a respective initial absolute relevance score above an absolute relevance score threshold;
determining an optimal promotion category tier-level for the consumer based on the consumer location and a total count of available promotions in the subset of available promotions;
generating an adjusted subset of available promotions based on the subset of available promotions and the optimal promotion category tier-level; and
transmitting the adjusted subset of available promotions to the computing device associated with the consumer.
18. The system of claim 17 , wherein the ranking target variable is associated with a probability that the consumer will purchase a particular promotion using the computing device or a probability that the consumer will click on an impression of a particular promotion using the computing device.
19. The system of claim 17 , further comprising:
ranking the adjusted subset of available promotions based on their respective initial absolute relevance scores.
20. The system of claim 19 , further comprising:
generating at least one personalized notification trigger for the consumer using the ranked adjusted subset of available promotions, the promotion data, and the user data.
21. The system of claim 17 , further comprising:
receiving initialization parameters including a target number of promotions to be selected, a minimum estimated absolute relevance probability, and a promotion category tier-level associated with the consumer location;
receiving a first subset of available promotions each associated with the promotion category tier-level of the multi-tier promotion categories;
calculating a first count of the first subset of available promotions that are associated with an absolute relevance score greater than or equal to the minimum estimated absolute relevance probability; and
in circumstances where the first count is greater than or equal to a scores count threshold, and in circumstances where the first count is less than or equal to the target number of promotions,
selecting the first subset of available promotions to be ranked.
22. The system of claim 21 , further comprising:
in circumstances where the first count is less than the scores count threshold,
increasing the promotion category tier-level to include a promotion subcategory of multi-tier the promotion categories;
receiving a second subset of available promotions each associated with the increased promotion category tier-level of the multi-tier promotion categories; and
calculating a second count of the second subset of available promotions that are associated with an absolute relevance score that is greater than or equal to the minimum estimated absolute relevance probability.
23. The system of claim 21 , further comprising:
in circumstances where the first count is greater than the target number of promotions,
decreasing the promotion category tier-level to exclude a promotion subcategory of the multi-tier promotion categories;
receiving a third subset of available promotions each associated with the decreased promotion category tier-level of the promotion category; and
calculating a third count of the third subset of available promotions that are associated with an absolute relevance score that is greater than or equal to the minimum estimated absolute relevance probability.
24. The system of claim 17 , wherein determining the optimal promotion category tier-level for the consumer includes determining a personalized promotion category multi-tier search for the consumer.