System and method for providing context-based third-party content

Patent metadata

Field	Value
Publication number	US-10182123-B2
Application number	US-201514662188-A
Country	US
Kind code	B2
Filing date	Mar 18, 2015
Priority date	Mar 18, 2015
Publication date	Jan 15, 2019
Grant date	Jan 15, 2019

Abstract

Systems and methods of the present disclosure can provide an improved user experience on client devices selecting to receive context-based third-party content by reducing the number of third-party content items presented for display on such client devices. A data processing system can be configured to employ an auction mechanism with selection criterion thresholds (such as minimum bid values) that result, when used in auctions, in enforcing a limit on the amount of third-party content presented for display on client devices that have selected to receive context-based third-party content. The selection criterion thresholds can result, when used in auctions, in enforcing minimum revenues for publishers. The selection criterion thresholds can be determined offline based on historical statistical parameters.

First claim

What is claimed is:

1. A method of providing third-party content, comprising:
determining, by a data processing system including at least one processor, at least one statistical parameter indicative of a joint probability distribution of a selection score among a plurality of publisher groups and a plurality of device groups, each publisher group of the plurality of publisher groups including one or more respective publishers and each device group of the plurality of device groups including one or more respective client computer devices that have selected to receive context-based third-party content, for each group pair including a device group of the plurality of device groups and a publisher group of the plurality of device groups, the joint probability distribution indicative of a probability of the selection score being associated with devices of the device group in the group pair and publishers of the publisher group in the group pair;
obtaining, by the data processing system, an indication of a predetermined threshold amount of third-party content for display to a client computer device in a first device group of the plurality of device groups;
obtaining, by the data processing system, a predetermined threshold publisher revenue value for a publisher in a first publisher group of the plurality of publisher groups;
determining, by the data processing system, a selection score threshold of the selection score specific to a group pair including the first device group and the first publisher group based on the at least one statistical parameter indicative of the joint probability distribution of the selection score, the predetermined threshold amount of third-party content for display to the client computer device, and the predetermined threshold publisher revenue value for the publisher;
receiving, by the data processing system, one or more bids for providing third-party content for display on the client computer device with content associated with the publisher, each bid associated with a respective third-party content item;
determining, by the data processing system, a bidding score for each third-party content item associated with a respective bid;
determining, by the data processing system, that at least one of the one or more bidding scores exceeds or is equal to the selection score threshold; and
selecting, by the data processing system, a third-party content item associated with a bidding score exceeding or equal to the selection score threshold as a candidate for display to the client computer device with the content associated with the publisher.

2. The method of claim 1 , wherein the one or more bids include one or more first bids associated with a first auction, the one or more third-party content items include one or more first third-party content items and the one or more bidding scores include one or more first bidding scores, the method comprising:
receiving, by the data processing system, one or more second bids associated with a second auction, for providing other third-party content for display on the client computer device with other content associated with the publisher, each second bid associated with a respective second third-party content item;
determining, by the data processing system, a second bidding score for each second third-party content item;
determining, by the data processing system, that none of the one or more second bidding scores exceeds or is equal to the selection score threshold; and
providing, by the data processing system, none of the one or more second content items for display on the client computer device with the other content associated with the publisher.

3. The method of claim 1 , wherein the selection score threshold includes a minimum bid value for each client device-publisher pair with a respective client device in the first device group and a respective publisher in the first publisher group.

4. The method of claim 1 , comprising:
obtaining, by the data processing system, for each device group of the plurality of device groups, an indication of a corresponding predetermined threshold amount of third-party content for display to client computer devices in that device group;
obtaining, by the data processing system, for each publisher group of the plurality of publisher groups, a corresponding predetermined threshold publisher revenue value for publishers in that publisher group; and
determining, for each group pair including a respective device group and a respective publisher group, a corresponding selection score threshold for client device-publisher pairs associated with the group pair based on the at least one statistical parameter, the corresponding predetermined threshold publisher revenue value for publishers in the respective publisher group and the corresponding predetermined threshold amount of third-party content for client computer devices in the respective device group.

5. The method of claim 4 , comprising:
determining the corresponding selection score threshold for each group pair by computing an estimate of the corresponding selection score threshold and iteratively:
identifying publishers in the respective publisher group for which expected revenues are below the corresponding predetermined threshold publisher revenue value for publishers in the respective publisher group;
adjusting the estimate of the corresponding selection score threshold for client device-publisher pairs associated with the identified publishers, the adjusted estimate causing the expected publisher revenues for the identified publishers to exceed the corresponding predetermined threshold publisher revenue value for publishers in the respective publisher group; and
re-computing the corresponding selection score threshold for client device-publisher pairs not associated with previously identified publishers.

6. The method of claim 1 , comprising:
maximizing weighted averages of economic efficiency for third-party content providers and publisher revenues.

7. The method of claim 1 , wherein the at least one statistical parameter includes a representation of a joint probability density function of the selection score.

8. The method of claim 1 , comprising:
obtaining, by the data processing system, data associated with previous bids made in association with each device group of the plurality of device groups and each publisher group of the plurality of publisher groups; and
computing the at least one statistical parameter based on the data associated with the previous bids.

9. A data processing system comprising:
a memory storing computer code instructions; and
a processor configured, upon executing the computer code instructions, to:
determine at least one statistical parameter indicative of a joint probability distribution of a selection score among a plurality of publisher groups and a plurality of device groups, each publisher group of the plurality of publisher groups including one or more respective publishers and each device group of the plurality of device groups including one or more respective client computer devices that have selected to receive context-based third-party content, for each group pair including a device group of the plurality of device groups and a publisher group of the plurality of device groups, the joint probability distribution indicative of a probability of the selection score being associated with devices of the device group in the group pair and publishers of the publisher group in the group pair;
obtain an indication of a predetermined threshold amount of third-party content for display to a client computer device in a first device group of the plurality of device groups;
obtain a predetermined threshold publisher revenue value for a publishe

Assignees

• Google Inc
• Google Llc

Inventors

• Hummel Patrick
• Varian Hal Ronald

Classifications

• G06Q30/0277

  Online advertisement · CPC title

• G06Q30/0275

  Auctions · CPC title

• G06Q30/0251 Primary

  Targeted advertisements · CPC title

• H04L67/06

  specially adapted for file transfer, e.g. file transfer protocol [FTP] · CPC title

• H04L67/20 Primary

  Electricity · mapped topic