System and method for synchronizing data on a network

• US 8,069,142 B2
• Filed: 12/06/2007
• Issued: 11/29/2011
• Est. Priority Date: 12/06/2007
• Status: Active Grant

• Current Assignee: Verizon Patent and Licensing Incorporated (Verizon Communications Inc.)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Grinstead, George, Trout, Robert Carter, Davis, Marc Eliot, Higgins, Christopher William, Martinez, Ronald, Boerries, Marco, Parry, Lee J., O'Sullivan, Joseph James
• Primary Examiner(s): Vital; Pierre
• Assistant Examiner(s): Nguyen; Loan T
• Application Number:
  US11/952,007
• Publication Number:
  US 20090150373A1
• Time in Patent Office: 1,454 Days
• Field of Search: None
• US Class Current: 707/610
• CPC Class Codes: G11B 27/105 of operating discs
• CPC Class Codes: H04L 67/1095 Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
• CPC Class Codes: H04L 67/125 involving control of end-device applications over a network
• CPC Class Codes: H04L 67/306 User profiles
• CPC Class Codes: H04N 21/258 Client or end-user data management, e.g. managing client capabilities, user preferences or demographics, processing of multiple end-users preferences to derive collaborative data arrangements for services using the result on the distributing side of broadcast systems H04H60/66; profiles in network data switching protocols H04L67/30

Abstract

The disclosure describes systems and methods for synchronizing data on a network based on temporal, spatial, social and logical data available to the network. The method includes receiving a first information object (IO) containing attributes for a first real-world entity (RWE), the first IO associated with a second RWE; identifying one or more second IOs, each second IO containing one or more attributes for the first RWE and each second IO independently associated with a third RWE; generating a different probability for each IO based on a comparison of contents of the first and second IOs and their associated RWEs; and replacing one or more of the attributes in at least one IO with at least one attribute from a different IO based on the probabilities for each IO.

Claims

1. A method comprising:

receiving, over a network, a first information object (IO) comprising for a target real-world entity (RWE), the first IO controlled by an owner RWE;

identifying, via a computing device, a plurality of second IOs, each of the plurality of second IOs comprising one or more attributes for the target RWE, each of the plurality of second IOs independently controlled by a different one of a plurality of third-party RWEs, the plurality of third-party RWEs not comprising the target RWE and the owner RWE;

retrieving social data, spatial data, temporal data and logical data available via the network associated with each RWE;

correlating, via the computing device, the first IO, the plurality of second IOs and the social data, spatial data, temporal data and logical data where a probability score is generated for individual attributes within each IO; and

automatically synchronizing, via the computing device, one or more of the attributes in at least one IO, said synchronization comprising replacing the one or more attributes in the at least one IO with a corresponding attribute from a different IO based on the probability score of the respective attribute.

2. The method of claim 1 further comprising:

comparing each attribute in the first IO with a set of most probable attributes for the target RWE; and

replacing one or more attributes in the first IO with a most probable attribute obtained from a second IO.

3. The method of claim 1 further comprising:

comparing each attribute in a selected second IO with a set of most probable attributes for the target RWE; and

replacing one or more attributes in the selected second IO with a most probable attribute obtained from a different IO.

4. The method of claim 1 such that where correlating comprises:

for each IO, determining a relationship between the target RWE and the IO'"'"'s associated owner or third-party RWE based on the at least one retrieved social data, spatial data, temporal data and logical data; and

generating the probability scores for each IO based on the relationship between the target RWE and the IO'"'"'s associated owner RWE or third-party RWE.

5. The method of claim 4, further comprising:

assigning each relationship a relative weight; and

generating the probability scores for each IO based at least in part on the relative weight assigned to the relationship between the target RWE and the IO'"'"'s associated owner RWE or third-party RWE.

6. The method of claim 4, where said determining a relationship comprises

comparing current attributes of the target RWE and the owner or third-party RWE;

comparing past attributes of the target RWE and the owner or third-party RWE;

retrieving a relationship previously selected by one of the target RWE or the owner or third-party RWE;

comparing attributes of third-party RWEs having known relationships with both the target RWE and the owner RWE; and

identifying messages between the target RWE and the owner or third-party RWE.

7. The method of claim 1, where the first IO includes a unique entity identifier for the target RWE and identifying a plurality of second IOs further comprises:

identifying a plurality of second IOs comprising attributes associated with the unique entity identifier for the target RWE.

8. The method of claim 1, where correlating comprises generating, via the computing device, a combined graph of the social data, spatial data, temporal data and logical data.

9. The method of claim 1, where the combined graph comprises a histogram.

10. The method of claim 1, where the social data, spatial data, temporal data and logical data comprise a feature vector comprising raw sensed data relating to RWEs and contextual and periodic patterns of the states and action of RWEs.

11. The method of claim 10, where correlating comprises processing the feature vector to determine similarity, difference, clustering, hierarchical and graph and inferential relationships between RWEs to which the social data, spatial data, temporal data and logical data within the feature vector relate.

12. A method comprising:

receiving, over a network, a first information object (IO) controlled by an owner RWE comprising a plurality of first IO attributes for a target real-world entity (RWE), the plurality of first IO attributes comprising social data, spatial data, temporal data and logical data, one of the plurality of first IO attributes comprising new information for the target RWE;

identifying, via the computing device, a plurality of third-party RWEs using the at least some of plurality of first IO attributes, where each of the plurality of third-party RWEs are associated with the new information;

identifying, via the computing device, for each of the plurality of third-party RWEs, a respective plurality of third-party IOs associated with the respective third-party RWE, each of the respective third-party IOs comprising a plurality of third-party IO attributes;

retrieving social data, spatial data, temporal data and logical data available via the network associated with the target RWE, the first IO, and the plurality of third-part RWEs and the respective pluralities of third-party IOs associated with each of the plurality of third-part RWEs;

correlating, via the computing device, the first IO, the respective pluralities of third-party IOs and the social data, spatial data, temporal data and logical data to generate a probability score for each of the first IO and the respective pluralities of third-party IOs, the probability score representing a likelihood that attributes within the respective IO reflect correct information;

identifying, via the computing device, within a group of IOs comprising the first IO and each of the respective pluralities of third-party IOs, a most probable IO having the highest probability score within the group of IOs;

replacing, via the computing device, information in the each of first IO and the respective pluralities of third-party IOs using information in the most probable IO.