System for determining and optimizing for relevance in match-making systems
Disclosed are methods and apparatus for automatically determining the relevance of matches between entities. A set of one or more indicators of relevance for each of a plurality of matches may be detected, where each of the plurality of matches exists between a first entity and a different one of a plurality of entities. Each set of one or more indicators of relevance indicates a degree of two-way interest for a corresponding one of the plurality of matches, the degree of two-way interest indicating both a degree of interest of the first entity in the corresponding one of the plurality of entities and a degree of interest of the corresponding one of the plurality of entities in the first entity. A probability of relevance of each of the plurality of matches may be determined based at least in part upon a corresponding set of one or more indicators of relevance. Each of the plurality of matches may be ranked based at least in part on the corresponding probability of relevance. A ranking function may be trained based upon the probability of relevance of each of the plurality of matches. The ranking function may subsequently be applied to identify and rank matches (e.g., in the absence of indicators of relevance).
1. A method, comprising:
detecting a first behavioral feature for a first potential match for a first entity and a second behavioral feature for a second potential match for a second entity, wherein the first behavioral feature indicates a degree of at least one-way interest in the first entity by a third entity, and the second behavioral feature indicates a degree of at least one-way interest in the second entity by the third entity;
determining a probability of relevance of the first and second potential matches based at least in part upon the first and second behavioral features, wherein the first behavioral feature pertains to a view of a profile of the first entity on a dating service, and the second behavioral feature pertains to a view of a profile of the second entity on the dating service, the profile of the first entity and the profile of the second entity included in a plurality of candidate profiles on the dating service;
training a machine-learned ranking model (i) using a subset of features defined by the candidate profiles only and the probability of relevance of each of the first and second potential matches as inputs, and (ii) by minimizing a total loss, at least in part based on the first entity and the second entity; and
applying the ranking model to rank a potential match for a fourth entity, based at least in part on a first feature vector indicating features of the profile of the first entity and a second feature vector indicating features of the profile of the second entity, wherein the applying is performed at least in part by partitioning a space of feature values into regions.
2. The method as recited in claim 1 , wherein the determining is based at least in part on features of the first entity or the second entity.
3. The method as recited in claim 1 , wherein the first and second behavioral features do not provide click feedback from the first entity, the second entity, or the third entity.
4. The method as recited in claim 1 , wherein the training is to predict features that correlate with relevance.
5. The method as recited in claim 1 , wherein the ranking model minimizes the total loss, based on a gradient descent method.
6. A non-transitory, computer-readable medium storing thereon computer-readable instructions, comprising:
instructions for detecting a first behavioral feature for a first potential match for a first entity and a second behavioral feature for a second potential match for a second entity, wherein the first behavioral feature indicates a degree of at least one-way interest in the first entity by a third entity, and the second behavioral feature indicates a degree of at least one-way interest in the second entity by the third entity;
instructions for determining a probability of relevance of the first and second potential matches based at least in part upon the first and second behavioral features, wherein the first behavioral feature pertains to a view of a profile of the first entity on a dating service, and the second behavioral feature pertains to a view of a profile of the second entity on the dating service, the profile of the first entity and the profile of the second entity included in a plurality of candidate profiles on the dating service;
instructions for training a machine-learned ranking model (i) using a subset of features defined by the candidate profiles only and the probability of relevance of each of the first and second potential matches as inputs, and (ii) by minimizing a total loss, at least in part based on the first entity and the second entity; and
instructions for applying the ranking model to rank a potential match for a fourth entity, based at least in part on a first feature vector indicating features of the profile of the first entity and a second feature vector indicating features of the profile of the second entity, wherein the applying is performed at least in part by partitioning a space of feature values into regions.
7. The medium as recited in claim 6 , wherein the determining is based at least in part on features of the first entity or the second entity.
8. The medium as recited in claim 6 , wherein the first and second behavioral features do not provide click feedback from the first entity, the second entity, or the third entity.
9. The medium as recited in claim 6 , wherein the training is to predict features that correlate with relevance.
10. The medium as recited in claim 6 , wherein the ranking model minimizes the total loss, based on a gradient descent method.
11. An apparatus, comprising:
a processor; and
a memory including instructions,
the processor, upon executing the instructions, configured for
detecting a first behavioral feature for a first potential match for a first entity and a second behavioral feature for a second potential match for a second entity, wherein the first behavioral feature indicates a degree of at least one-way interest in the first entity by a third entity, and the second behavioral feature indicates a degree of at least one-way interest in the second entity by the third entity,
determining a probability of relevance of the first and second potential matches based at least in part upon the first and second behavioral features, wherein the first behavioral feature pertains to a view of a profile of the first entity on a dating service, and the second behavioral feature pertains to a view of a profile of the second entity on the dating service, the profile of the first entity and the profile of the second entity included in a plurality of candidate profiles on the dating service,
training a machine-learned ranking model (i) using a subset of features defined by the candidate profiles only and the probability of relevance of each of the first and second potential matches as inputs, and (ii) by minimizing a total loss, at least in part based on the first entity and the second entity, and
applying the ranking model to rank a potential match for a fourth entity, based at least in part on a first feature vector indicating features of the profile of the first entity and a second feature vector indicating features of the profile of the second entity, wherein the applying is performed at least in part by partitioning a space of feature values into regions.
12. The apparatus as recited in claim 11 , wherein the determining is based at least in part on features of the first entity or the second entity.
13. The apparatus as recited in claim 11 , wherein the first and second behavioral features do not provide click feedback from the first entity, the second entity, or the third entity.
14. The apparatus as recited in claim 11 , wherein the ranking model minimizes the total loss, based on a gradient descent method.
15. The apparatus as recited in claim 11 , wherein the ranking model ranks the potential match for the fourth entity, based at least in part on a vector of the fourth entity and at least one of the regions.