Methods and apparatuses for trust computation

• US 9,444,846 B2
• Filed: 06/19/2014
• Issued: 09/13/2016
• Est. Priority Date: 06/19/2014
• Status: Expired due to Fees

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Sharma, Avinash, Dutta, Partha, Mukherjee, Tridib, Kunde, Shruti
• Primary Examiner(s): Shiferaw, Eleni
• Assistant Examiner(s): LAPIAN, ALEXANDER R
• Application Number:
  US14/308,748
• Publication Number:
  US 20150373049A1
• Time in Patent Office: 817 Days
• US Class Current: 1/1
• CPC Class Codes: H04L 63/126 the source of the received data
• CPC Class Codes: H04L 63/20 for managing network security; network security policies in general filtering policies H04L63/0227

Abstract

Embodiments of a method are disclosed for computing trust index among multiple entities associated with a resource marketplace. The method includes receiving multiple inputs including interaction attributes, attribute importance factors, references to databases, and multiple entities associated with the resource marketplace. The method also includes creating a weighted-interaction graph based on the received inputs. The weighted-interaction graph includes multiple vertices representing the entities. The method further includes performing a topology-specific analysis of the weighted-interaction graph. The method furthermore includes computing Euclidean distances for each pair of vertices in the weighted-interaction graph based on the performed analysis. The method also includes determining a trust index for a first entity in the received multiple entities based on the computed Euclidean distances. The trust index includes ranking of at least one of the multiple entities with respect to the first entity. The ranking is inversely proportional to the computed Euclidean distances.

Claims

1. A method for computing trust index among a plurality of entities associated with a resource marketplace, the method comprising:

receiving a plurality of inputs including interaction attributes, attribute importance factors, references to databases, and a plurality of entities associated with the resource marketplace, wherein the entities comprise at least users, resources, and resource owners;

creating a weighted-interaction graph based on the received plurality of inputs, wherein the weighted-interaction graph includes a plurality of vertices representing the plurality of entities;

performing a topology-specific analysis of the weighted-interaction graph;

computing Euclidean distances for each pair of vertices among the plurality of vertices in the weighted-interaction graph based on the performed analysis; and

determining a trust index for a first entity in the received plurality of entities based on the computed Euclidean distances, the trust index includes ranking of at least one of the plurality of entities with respect to the first entity, the ranking being inversely proportional to the computed Euclidean distances, wherein

the first entity comprises a user and the trust index is used to determine one or more marketplace entities comprising resources for performing the user'"'"'s task, or

the first entity comprises a resource owner and the trust index is used to determine one or more marketplace entities comprising resources for performing a task received by the resource owner from a marketplace entity comprising a user.

2. The method of claim 1, further comprising determining at least one community of entities within the plurality of entities based on at least one of the interaction attributes and the determined trust index.

3. The method of claim 2, wherein the trust index for the first entity is determined based on a combination of the computed Euclidean distances and the determined at least one community of entities to which the first entity is associated.

4. The method of claim 1, wherein the interaction attributes include at least one of social network connectivity, geographical location, organizational affiliation, hardware profile, usage frequency, and user feedback for the plurality of entities.

5. The method of claim 1, wherein the databases store at least one of user profiles, social network connectivity profiles, historical usage profiles, resource profiles, and resource owner profiles for the plurality of entities.

6. The method of claim 1, wherein performing the topology-specific analysis includes embedding the weighted-interaction graph in a spectral space of a dimension being predefined or received from at least one of the plurality of entities.

7. The method of claim 6, wherein the dimension is smaller than the total number of vertices in the weighted-interaction graph.

8. The method of claim 1, wherein the weighted-interaction graph is undirected in nature.

9. The method of claim 1, wherein interactions within the plurality of entities are weighted using predetermined weighting functions.

10. The method of claim 9, wherein at least one of the weighting functions is a linear combination of one or more of the interaction attributes and the attribute importance factors.

11. The method of claim 9, wherein a sum of the attribute importance factors associated with each of the predetermined weighting functions is one.

12. The method of claim 9, wherein values of the weighting functions are normalized to range from zero to one.

13. The method of claim 9, wherein values of the weighting functions are computed periodically or upon request.

14. The method of claim 1, wherein values of the interaction attributes are computed according to predetermined relationship criteria for the plurality of entities.

15. The method of claim 1, wherein at least one of the steps of creating, performing, computing, and determining is embodied in software that is configured for implementation by a processor.

16. The method of claim 15, wherein the step of receiving a plurality of inputs includes receiving inputs that are configured for use with an input/output device, such that the input/output device is configured to:

receive the inputs relating to the interaction attributes, the attribute importance factors, the references to the databases, and the plurality of entities associated with the resource marketplace; and

output corresponding data to the processor.

17. The method of claim 1, wherein the step of receiving the plurality of inputs includes receiving the inputs via a network.

20. The method of claim 2, wherein the determined community of entities comprises at least users and resource owners, and the community of entities is determined based on embedding the weighted-interaction graph in a K-dimensional spectral space and performing K-means clustering on the K-dimensional spectral space.

18. A device for computing trust index for at least one entity interacting with a plurality of entities associated with a resource marketplace, the device comprising:

a memory;

a database including profiles and historical usage information for the plurality of entities, wherein the device is further configured to;

receive a plurality of inputs including at least one of the plurality of entities, interaction attributes, attribute importance factors for each of the interaction attributes, and a reference to the database; and

create a weighted-interaction graph based on the plurality of inputs, the integration graph includes vertices representing the at least one entity and the plurality of entities, wherein the entities comprise at least users, resources, and resource owners;

perform a topology-specific analysis of the weighted-interaction graph;

compute Euclidean distances for each pair of vertices in the weighted-interaction graph based on the performed topology-specific analysis; and

determine a trust index for the at least one entity based on the computed Euclidean distances, the trust index includes ranking of at least one of the plurality of entities with respect to the at least one entity, the ranking being inversely proportional to the computed Euclidean distances, wherein

the first entity comprises a user and the trust index is used to determine one or more marketplace entities comprising resources for performing the user'"'"'s task, or

the first entity comprises a resource owner and the trust index is used to determine one or more marketplace entities comprising resources for performing a task received by the resource owner from a marketplace entity comprising a user.

19. A non-transitory computer-readable medium comprising computer-executable instructions for computing trust index among a plurality of entities associated with a resource marketplace, the non-transitory computer-readable medium comprising instructions for:

receiving a plurality of inputs including interaction attributes, attribute importance factors, instances of accessible databases, and a plurality of entities associated with the resource marketplace;

creating a weighted-interaction graph based on the received plurality of inputs, wherein the weighted-interaction graph includes vertices being represented by the plurality of entities, wherein the entities comprise at least users, resources, and resource owners;

performing a topology-specific analysis of the weighted-interaction graph;

computing Euclidean distances for each pair of vertices in the weighted-interaction graph based on the performed analysis; and

determining a trust index for a first entity in the received plurality of entities based on the computed Euclidean distances, the trust index includes ranking of at least one of the plurality of entities with respect to the first entity, the ranking being inversely proportional to the computed Euclidean distances, wherein

the first entity comprises a user and the trust index is used to determine one or more marketplace entities comprising resources for performing the user'"'"'s task, or

the first entity comprises a resource owner and the trust index is used to determine one or more marketplace entities comprising resources for performing a task received by the resource owner from a marketplace entity comprising a user.