Back to results
US 11,748,776 B2Granted
Systems and methods of generating context specification for contextualized searches and content delivery
AssigneeStackadapt, Inc.
InventorNedialko Dimitrov
Life of this patent
Estimating the term…
Abstract
Systems and methods described herein may generate campaigns and efficiently calculate bids for placement of campaign data into Internet data. Embodiments may calculate context scores for campaign data based on campaign terms and beacon terms. The context scores may be used to identify Internet content that has a high page score. If the page score of particular Internet content exceeds a predetermined threshold, the system may place a bid for a campaign based on disclosed algorithms taking as inputs performance scores, context scores, page scores, campaign budgets, and other parameters. The systems therefore are capable of quickly and effectively calculating optimal bids to place for a particular campaign given parameters disclosed herein.
Claims20
A computer-implemented method comprising:
- prior to a bid-time and after training a machine-learning model,
- applying, by a computer, the machine-learning model to a first set of context terms received from a client device to output a set of beacon terms extracted from a plurality of corpus terms stored in a corpus database, wherein the machine-learning model is trained on the plurality of corpus terms stored in the corpus database, wherein the machine-learning model determines a plurality of co-occurrence probabilities corresponding to the plurality of corpus terms indicating a probability that one or more context terms co-occurs with one or more corpus terms, and wherein the machine-learning model extracts each beacon term from each corpus term satisfying a threshold co-occurrence probability;
- calculating, by the computer, a plurality of page scores for a plurality of corpus webpages stored in the corpus database based upon the set of beacon terms satisfying the threshold co-occurrence probability and the first set of context terms;
- identifying, by the computer, in the plurality of corpus webpages a set of contextual webpages having page scores satisfying a threshold;
- applying, by the computer, the machine-learning model to the first set of context terms and a second set of context terms received from the client device to output an updated set of beacon terms;
- calculating, by the computer, one or more updated page scores for one or more corpus webpages stored in the corpus database based upon the updated set of beacon terms, the first set of context terms, and the second set of context terms;
- updating, by the computer, the set of contextual webpages based upon the one or more updated page scores;
- and
- storing, by the computer into a campaign database, campaign data of a user comprising the set of beacon terms and the set of contextual webpages, the campaign data configured for executing at a future bid-time a real-time bidding selection operation for one or more available webpages during the real-time bidding selection operation.
6 more elements
A system comprising:
- a corpus database comprising non-transitory storage medium configured to store at least a portion of a plurality of Internet content data corresponding to a uniform resource locator (URL) and page context score;
- a campaign database comprising non-transitory storage medium configured to store campaign data for a plurality of users, the campaign data configured for executing a real-time bidding selection operation for one or more available webpages during the real-time bidding selection operation;
- and
- a server comprising a processor configured to:
- at a bid time and after training a machine-learning model,
- apply the machine-learning model to a first set of context terms received from a client device to output a set of beacon terms extracted from a plurality of corpus terms stored in the corpus database, wherein the machine-learning model is trained on the plurality of corpus terms stored in the corpus database, wherein the machine-learning model determines a plurality of co-occurrence probabilities corresponding to the plurality of corpus terms indicating a probability that one or more context terms co-occurs with one or more corpus terms, and wherein the machine-learning model extracts each beacon term from each corpus term satisfying a threshold co-occurrence probability;
- calculate a plurality of page scores for a plurality of corpus webpages stored in the corpus database based upon the set of beacon terms satisfying the threshold co-occurrence probability and the first set of context terms;
- identify in the plurality of corpus webpages a set of contextual webpages having page scores satisfying a threshold;
- apply the machine-learning model to the first set of context terms and a second set of context terms received from the client device to output an updated set of beacon terms;
- calculate one or more updated page scores for one or more corpus webpages stored in the corpus database based upon the updated set of beacon terms, the first set of context terms, and the second set of context terms;
- update the set of contextual webpages based upon the one or more updated page scores;
- and
- store, into the campaign database, campaign data of a user comprising the set of beacon terms and the set of contextual webpages, the campaign data configured for executing at a future bid-time a real-time bidding selection operation for one or more available webpages during the real-time bidding selection operation.
10 more elements
A computer readable medium containing machine-executable program instructions, wherein execution of the program instructions by one or more processors of a computer system causes the one or more processors to execute the steps of:
- prior to a bid-time and after training a machine-learning model,
- applying the machine-learning model to a first set of context terms received from a client device to output a set of beacon terms extracted from a plurality of corpus terms stored in a corpus database, wherein the machine-learning model is trained on the plurality of corpus terms stored in the corpus database, wherein the machine-learning model determines a plurality of co-occurrence probabilities corresponding to the plurality of corpus terms indicating a probability that one or more context terms co-occurs with one or more corpus terms, and wherein the machine-learning model extracts each beacon term from each corpus term satisfying a threshold co-occurrence probability;
- calculating a plurality of page scores for a plurality of corpus webpages stored in the corpus database based upon the set of beacon terms satisfying the threshold co-occurrence probability and the first set of context terms;
- identifying in the plurality of corpus webpages a set of contextual webpages having page scores satisfying a threshold;
- applying the machine-learning model to the first set of context terms and a second set of context terms received from the client device to output an updated set of beacon terms;
- calculating one or more updated page scores for one or more corpus webpages stored in the corpus database based upon the updated set of beacon terms, the first set of context terms, and the second set of context terms;
- updating the set of contextual webpages based upon the one or more updated page scores;
- and
- storing, into a campaign database, campaign data of a user comprising the set of beacon terms and the set of contextual webpages, the campaign data configured for executing at a future bid-time a real-time bidding selection operation for one or more available webpages during the real-time bidding selection operation.
6 more elements
Dependencies are inferred from the claim text.