COLLECTIONS OF LINKED DATABASES

• US 20140136522A1
• Filed: 01/20/2014
• Published: 05/15/2014
• Est. Priority Date: 09/15/2004
• Status: Active Grant

• Current Assignee: 3degrees LLC
• Original Assignee: 3degrees LLC
• Inventors: Markus, Michael J., Markus (nee McGuire), Heather A., Kionga-Kamau, Peter M., Smith, Brian N.
• Granted Patent: US 10,733,242 B2
• Time in Patent Office: Days
• US Class Current: 707/722
• CPC Class Codes: G06F 16/906 Clustering; Classification
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9536 Search customisation based on social or collaborative filtering

Abstract

A method and system generate an inferential network. Records of searches of a network of linked nodes are stored. Each of the nodes is connected to another one of the nodes by a link. Further, each of the records includes: 1) search terms used in a corresponding search; and 2) results of the corresponding search. The results correspond to nodes of the network. The stored records are analyzed to infer relationships between pairs of nodes of the network. Each of the relationships is inferred from the behavior of one of the nodes of a pair of nodes relative to the other node of the pair of nodes. Inferential nodes corresponding to the nodes of the pairs of nodes are generated, and inferential links between the inferential nodes are generated. The inferential links correspond to the inferred relationships.

Claims

1. A method for generating an inferential network, the method comprising:

storing records of searches of a network of linked nodes, each of the nodes connected to another one of the nodes by a link, and each of the records including;

search terms used in a corresponding search; and

results of the corresponding search, the results corresponding to nodes of the network;

analyzing the stored records to infer relationships between pairs of nodes of the network, each of the relationships inferred from the behavior of one of the nodes of a pair of nodes relative to the other node of the pair of nodes;

generating inferential nodes corresponding to the nodes of the pairs of nodes; and

generating inferential links between the inferential nodes, the inferential links corresponding to the inferred relationships;

wherein the method is performed by at least one processor.

2. The method of claim 1, wherein each of the inferential links extends between the inferential nodes of a pair of inferential nodes, the pair of inferential nodes corresponding to the pair of nodes of the inferred relationship corresponding to the inferential link.

3. The method of claim 1, wherein one of the pairs of nodes corresponding to one of the records includes:

a first node corresponding to a member of the network, the member being the initiator of the corresponding search of the record; and

a second node selected by the member as being;

most likely to possess information on a topic of the corresponding search of the record;

or

most likely to be directly linked to another node of the network possessing information on the topic.

4. The method of claim 1, wherein the nodes of the network correspond to one or more of members of a social network and websites.

5. The method of claim 1, further including:

generating one of the inferential links based on one of the stored records; and

labeling the inferential link based on search terms of the stored record.

6. The method of claim 1, further including:

labeling one of the inferential links with a strength.

7. The method of claim 6, further including:

decreasing the strength over time.

8. The method of claim 6, further including:

increasing the strength with the number of times that a node of the network corresponding to the inferential link is designated as an anchor database in a search.

9. The method of claim 1, further including:

determining an inferential network path from a first inferential node of the inferential network to a second inferential node of the inferential network using existing inferential links of the inferential network.

10. The method of claim 9, wherein the inferential links include strengths and the method further includes:

determining the inferential network path as a function of the strengths.

11. The method of claim 1, further including:

displaying at least a portion of the inferential network having the form of;

D₁-c-D₂,

wherein “

D1”

is a first inferential node, “

D2”

is a second inferential node, “

-”

represents an inferential link between D1 and D2, and “

c”

represents a visual designation of the “

-”

link between the first inferential node and the second inferential node, the visual designation of the “

-”

link visually qualifying the link and visually differing from a visual designation of another “

-”

link of the displayed portion of the inferential network.

12. The method of claim 11, wherein “

c”

is based at least in part upon a strength of the inferential link between D1 and D2.

13. A system for generating an inferential network, the system comprising:

at least one memory storing records of searches of a network of linked nodes, each of the nodes connected to another one of the nodes by a link, and each of the records including;

search terms used in a corresponding search; and

results of the corresponding search, the results corresponding to nodes of the network;

at least one program memory programmed to;

analyze the stored records to infer relationships between pairs of nodes of the network, each of the relationships inferred from the behavior of one of the nodes of a pair of nodes relative to the other node of the pair of nodes;

generate inferential nodes corresponding to the nodes of the pairs of nodes; and

generate inferential links between the inferential nodes, the inferential links corresponding to the inferred relationships;

14. The system of claim 13, wherein each of the inferential links extends between the inferential nodes of a pair of inferential nodes, the pair of inferential nodes corresponding to the pair of nodes of the inferred relationship corresponding to the inferential link.

15. The system of claim 13, wherein one of the pairs of nodes corresponding to one of the records includes:

a first node corresponding to a member of the network, the member being the initiator of the corresponding search of the record; and

a second node selected by the member as being;

most likely to possess information on a topic of the corresponding search of the record;

or

most likely to be directly linked to another node of the network possessing information on the topic.

16. The system of claim 13, wherein the at least one processor is further programmed to:

generate one of the inferential links based on one of the stored records; and

label the inferential link based on search terms of the stored record.

17. The system of claim 16, wherein the at least one processor is further programmed to:

label one of the inferential links with a strength.

18. The system of claim 16, wherein the at least one processor is further programmed to:

increase the strength with the number of times that a node of the network corresponding to the inferential link is designated as an anchor database in a search.

19. The system of claim 13, wherein the at least one processor is further programmed to:

determine an inferential network path from a first inferential node of the inferential network to a second inferential node of the inferential network using existing inferential links of the inferential network.

20. The method of claim 13, wherein the at least one processor is further programmed to:

display at least a portion of the inferential network having the form of;

D₁-c-D₂,

wherein “

D1”

is a first inferential node, “

D2”

is a second inferential node, “

-”

represents an inferential link between D1 and D2, and “

c”

represents a visual designation of the “

-”

link between the first inferential node and the second inferential node, the visual designation of the “

-”

link visually qualifying the link and visually differing from a visual designation of another “

-”

link of the displayed portion of the inferential network.