Prior art — US Patent 10134054

Analysis of Prior Art for U.S. Patent 10,134,054

Report Date: April 29, 2026

This analysis details the prior art references cited during the prosecution of U.S. Patent 10,134,054 ("the '054 patent"). The '054 patent has a priority date of June 16, 2006, derived from provisional application 60/804,937. Therefore, any reference published or filed before this date qualifies as prior art under 35 U.S.C. § 102.

The primary claims of the '054 patent revolve around a first computer system (e.g., a "BT company") that collects user profile information, and a second, separate computer system (e.g., an "ad network") that controls ad space on various websites. A key inventive concept is the first system causing the second system to "tag" a user's device and providing a "condition" for serving an ad, all without transferring the sensitive user profile data to the second system. The ad is only served after the user is redirected back to the first system.

The following references were cited by the USPTO examiner during prosecution and are listed on the face of the patent.

Cited References and Potential Anticipation

1. U.S. Patent No. 7,747,745 (Shkedi)

Full Citation: U.S. Patent No. 7,747,745, "Media properties selection method and system based on expected profit from profile-based ad delivery," issued June 29, 2010.
Filing Date: June 14, 2007. (This patent claims priority to the same 2006 provisional application as the '054 patent, making it part of the same patent family. It is cited for informational purposes but is not prior art that can anticipate the claims under § 102).
Brief Description: This patent, by the same inventor, describes a system for selecting which media properties (websites) to target with ads based on a calculation of expected profit. A Behavioral Targeting (BT) company collects a user's profile and calculates the potential revenue versus the cost of ad space on various properties. If the profit is positive, it arranges for the user to be tagged in a way that the selected media property can read.
Potential Anticipation Analysis: As this is a continuation of the same inventive effort and shares the same priority date, it does not anticipate the '054 patent. It does, however, provide foundational context for the invention, describing the profit-calculation and selection method that precedes the specific conditional, privacy-sensitive interaction with a second computer system detailed in the '054 claims.

2. U.S. Patent No. 8,204,783 (Shkedi)

Full Citation: U.S. Patent No. 8,204,783, "Media properties selection method and system based on expected profit from profile-based ad delivery," issued June 19, 2012.
Filing Date: June 25, 2010. (This is also part of the same patent family as the '054 patent and is not valid prior art against it).
Brief Description: Like the '745 patent, this continuation further details the method of selecting media properties for ad delivery based on profitability. It elaborates on how a BT company can redirect a visitor to a selected media property to have that property tag the user with its own tag (e.g., a cookie under the media property's domain).
Potential Anticipation Analysis: This patent is not prior art to the '054 patent. It is part of the claimed invention's lineage.

3. U.S. Patent Application Publication No. 2005/0198007 (Schwartz et al.)

Full Citation: US 2005/0198007 A1, "Method and system for brokering user profile data," published September 8, 2005.
Filing Date: March 5, 2004. (This is valid prior art).
Brief Description: Schwartz describes a system for brokering user profile data between data providers and data consumers (e.g., advertisers). It includes a central profile data broker that manages user profiles and allows advertisers to query for users matching certain criteria. A key aspect is that the broker can control the release of information and can serve ads to targeted users without necessarily revealing the full profile to the advertiser. It describes using a user ID to anonymously identify users across different sites.
Potential Anticipation Analysis:
- Claims 1, 6, 11: Schwartz discloses a system with separate entities (data providers, brokers, advertisers) and the use of anonymous IDs (tags) to target users without revealing their full profiles. This arguably teaches several elements of the '054 claims, such as tagging a user and serving an ad based on a profile held by a different entity. However, a key distinction and the likely reason the '054 patent was granted over this reference is the specific flow claimed in '054: the first system causes the second system to tag the user and transfers a specific condition to that second system, which later triggers a redirection back to the first system for the ad serving event. Schwartz's broker system appears more centralized, potentially serving the ad itself or providing data directly to the advertiser under certain rules, which differs from the specific redirection-based method of the '054 claims. The "transferring a condition" step is a particularly strong point of novelty for the '054 patent over Schwartz.

4. U.S. Patent No. 6,925,441 (Ambrose et al.)

Full Citation: U.S. Patent No. 6,925,441, "Distributed data capability network with client data validation and server data verification," issued August 2, 2005.
Filing Date: December 29, 2000. (This is valid prior art).
Brief Description: Ambrose discloses a distributed network where advertisers can deliver targeted content to users. It describes a "targeting service" that stores user profiles. When a user visits a publisher's site, the publisher can request an ad from an ad server, which in turn queries the targeting service. The targeting service can return an ad or a segment ID for the user without revealing the underlying profile data to the publisher or ad server.
Potential Anticipation Analysis:
- Claims 1, 6, 11: Ambrose teaches the separation of the profile-holding entity (targeting service) from the ad-serving entity. It also teaches the concept of targeting without transferring the full profile. However, like Schwartz, it does not appear to explicitly teach the claimed sequence of the first system instructing a second, unaffiliated ad network to tag a user, providing a specific condition for that tag, and then having the second network redirect the user back to the first system for ad serving only when that condition is met. The specific, condition-based trigger for the redirection is a critical element of the '054 claims that appears to be absent in Ambrose.

5. U.S. Patent No. 7,133,851 (Fay et al.)

Full Citation: U.S. Patent No. 7,133,851, "System and method for targeting advertisements," issued November 7, 2006.
Filing Date: May 19, 2000. (This is valid prior art).
Brief Description: This patent, assigned to DoubleClick, describes a core process of behavioral ad targeting. An ad server places a cookie on a user's browser during a visit to a website. This cookie contains or points to a user profile (e.g., interest categories). When the user later visits another site in the network, the ad server reads the cookie and serves an ad based on the stored profile information. This reference is foundational to the concept of retargeting.
Potential Anticipation Analysis:
- Claims 1, 6, 11: Fay describes a single system (the ad network/ad server) that both holds the profile information and controls the ad space. This is fundamentally different from the architecture in the '054 claims, which explicitly requires a first computer system (controlled by a first entity) and a second computer system not controlled by the first entity. The '054 patent's invention lies in the interaction between these two separate entities in a privacy-preserving manner, a distinction that renders Fay's disclosure non-anticipatory.

6. U.S. Patent No. 7,599,853 (Horvitz et al.)

Full Citation: U.S. Patent No. 7,599,853, "Privacy-sensitive personalization," issued October 6, 2009.
Filing Date: December 19, 2003. (This is valid prior art).
Brief Description: This Microsoft patent describes methods for personalizing content while preserving user privacy. It proposes storing user profile information locally on the user's machine. A service can send rules or models to the local machine, which then uses them to select content or ads based on the private local profile, without the profile data ever leaving the user's device.
Potential Anticipation Analysis:
- Claims 1, 6, 11: Horvitz's approach is quite different from the '054 patent. It focuses on client-side processing to maintain privacy, where the profile data remains with the user. The '054 patent, in contrast, describes a server-side architecture where the first entity's server maintains the profile. While both are "privacy-sensitive," the technical implementation is distinct. Horvitz does not describe the interaction between two separate server systems where one causes the other to tag a user and then apply a condition for a server-to-server redirection. Therefore, it does not anticipate the claims.