Method and system for matching entities in an auction
DCFirst Claim
1. A method of pairing requests, comprising:
- estimating at least one content-specific or requestor-specific characteristic associated with a request;
determining a set of available partners, each having at least one respective partner characteristic;
evaluating, with at least one automated processor, a plurality of pairings of the request with a plurality of different available partners, according to an evaluator for valuing pairings of the request with respective available partners, comprising at least one of a probabilistic function, a hierarchical Markov model, Bayesian logic, and a neural network, dependent on the at least one content-specific or requestor-specific characteristic, and the at least one respective partner characteristic; and
generating a control signal, by the at least one automated processor, selectively dependent on the evaluating.
1 Assignment
Litigations
0 Petitions
Accused Products
Abstract
A method for matching a first entity with at least one second entity selected from a plurality of second entities, comprising defining a plurality of multivalued scalar data representing inferential targeting parameters for the first entity and a plurality of multivalued scalar data of each of the plurality of second entities, representing respective characteristic parameters for each respective second entity; and performing an automated optimization with respect to an economic surplus of a respective match of the first entity with the at least one of the plurality of second entities, and an opportunity cost of the unavailability of the at least one of the plurality of second entities for matching with an alternate first entity.
475 Citations
20 Claims
-
1. A method of pairing requests, comprising:
-
estimating at least one content-specific or requestor-specific characteristic associated with a request; determining a set of available partners, each having at least one respective partner characteristic; evaluating, with at least one automated processor, a plurality of pairings of the request with a plurality of different available partners, according to an evaluator for valuing pairings of the request with respective available partners, comprising at least one of a probabilistic function, a hierarchical Markov model, Bayesian logic, and a neural network, dependent on the at least one content-specific or requestor-specific characteristic, and the at least one respective partner characteristic; and generating a control signal, by the at least one automated processor, selectively dependent on the evaluating. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18)
-
-
19. A system for pairing requests, comprising:
-
at least one port, configured to receive a request and information identifying a content or requestor-related characteristic associated with the request; at least one memory, configured to store availability information for a plurality of partners, and at least one respective partner characteristic associated with each respective available partner; and at least one automated processor, configured to generate a control signal, selectively dependent on an evaluation of each of a plurality of pairings of the request with a plurality of different available partners, dependent on at least one probabilistic function of the at least one content or requestor-related characteristic associated with the request, and the at least one respective partner characteristic.
-
-
20. A method of processing requests, comprising:
-
estimating at least one content-specific or requestor-specific characteristic associated with each received request; determining availability of a plurality of alternate target resources, each respective target resource having at least one respective target characteristic; evaluating, with at least one automated processor, a plurality of alternate allocations of the respective received request with different available targets, according to a ranking dependent on a probabilistic predictive multivariate evaluator, based on the at least one content-specific or requestor-specific characteristic, and the respective target characteristics of the plurality of alternate target resources; and generating a control signal, by the at least one automated processor, selectively dependent on the evaluating, to control the allocations of the respective received request with the different available targets.
-
Specification