User-powered recommendation system

Patent metadata

Field	Value
Publication number	US-8943081-B2
Application number	US-61689209-A
Country	US
Kind code	B2
Filing date	Nov 12, 2009
Priority date	Nov 17, 2008
Publication date	Jan 27, 2015
Grant date	Jan 27, 2015

Abstract

Recommendation systems are widely used in Internet applications. In current recommendation systems, users only play a passive role and have limited control over the recommendation generation process. As a result, there is often considerable mismatch between the recommendations made by these systems and the actual user interests, which are fine-grained and constantly evolving. With a user-powered distributed recommendation architecture, individual users can flexibly define fine-grained communities of interest in a declarative fashion and obtain recommendations accurately tailored to their interests by aggregating opinions of users in such communities. By combining a progressive sampling technique with data perturbation methods, the recommendation system is both scalable and privacy-preserving.

First claim

The invention claimed is:

1. A method for generating a recommendation, the method comprising:
receiving a user-defined recommendation request issued by a user as a user query;
receiving a user-defined declarative community definition;
selecting a sampling group belonging to a community defined by the declarative community definition;
for each particular member in the sampling group, retrieving, from a recommendation database, a particular rating data of each particular member based on the user-defined recommendation request;
for each particular rating data, generating a particular perturbed rating data, such that there is a predetermined probability that the particular perturbed rating data is different from a respective particular rating data;
aggregating all the particular perturbed rating data into a plurality of perturbed rating data; and
generating the recommendation based on the plurality of perturbed rating data.

2. The method of claim 1 , further comprising:
receiving a user-defined item constraint; and
generating the recommendation based on the user-defined item constraint.

3. The method of claim 1 , wherein the user-defined declarative community definition comprises a predicate on user attributes.

4. The method of claim 1 , wherein:
the community is split into a plurality of disjoint sampling groups;
each particular disjoint sampling group in the plurality of disjoint sampling groups has a particular size;
particular sizes decrease exponentially from a first size to a second size; and
the sampling group selected is a disjoint sampling group with a smallest size such that an accuracy level of the plurality of perturbed rating data is one of greater than and equal to a user-defined accuracy level.

5. The method of claim 1 , wherein selecting a sampling group comprises:
splitting the community into a plurality of disjoint sampling groups, wherein each particular disjoint sampling group in the plurality of disjoint sampling groups has a particular size, and wherein particular sizes decrease exponentially from a largest size to a smallest size; and
selecting a particular disjoint sampling group based on a user-defined sampling rate.

6. The method of claim 1 , further comprising:
receiving a user-defined recommendation algorithm; and
generating the recommendation based on the user-defined recommendation algorithm.

7. An apparatus for generating a recommendation, the apparatus comprising:
a processor; and
a memory to store computer program instructions, the computer program instructions when executed on the processor cause the processor to perform operations comprising:
receiving a user-defined recommendation request issued by a user as a user query;
receiving a user-defined declarative community definition;
selecting a sampling group belonging to a community defined by the declarative community definition;
for each particular member in the sampling group, retrieving, from a recommendation database, a particular rating data of each particular member based on the user-defined recommendation request;
for each particular rating data, generating a particular perturbed rating data, such that there is a predetermined probability that the particular perturbed rating data is different from a respective particular rating data;
aggregating all the particular perturbed rating data into a plurality of perturbed rating data; and
generating the recommendation based on the plurality of perturbed rating data.

8. The apparatus of claim 7 , the operations further comprising:
receiving a user-defined item constraint; and
generating the recommendation based on the user-defined item constraint.

9. The apparatus of claim 7 , wherein:
the community is split into a plurality of disjoint sampling groups;
each particular disjoint sampling group in the plurality of disjoint sampling groups has a particular size;
particular sizes decrease exponentially from a first size to a second size; and
the sampling group selected is a disjoint sampling group with a smallest size such that an accuracy level of the plurality of perturbed rating data is one of greater than and equal to a user-defined accuracy level.

10. The apparatus of claim 7 , the operations further comprising:
receiving a user-defined recommendation algorithm; and
generating the recommendation based on the user-defined recommendation algorithm.

11. The apparatus of claim 7 , wherein the selecting a sampling group comprises:
splitting the community into a plurality of disjoint sampling groups, wherein each particular disjoint sampling group in the plurality of disjoint sampling groups has a particular size, and wherein particular sizes decrease exponentially from a largest size to a smallest size; and
selecting a particular disjoint sampling group based on a user-defined sampling rate.

12. A non-transitory computer readable medium storing computer program instructions for generating a recommendation, which, when executed on a processor, cause the processor to perform a method comprising:
receiving a user-defined recommendation request issued by a user as a user query;
receiving a user-defined declarative community definition;
selecting a sampling group belonging to a community defined by the declarative community definition;
for each particular member in the sampling group, retrieving, from a recommendation database, a particular rating data of each particular member based on the user-defined recommendation request;
for each particular rating data, generating a particular perturbed rating data, such that there is a predetermined probability that the particular perturbed rating data is different from a respective particular rating data;
aggregating all the particular perturbed rating data into a plurality of perturbed rating data; and
generating the recommendation based on the plurality of perturbed rating data.

13. The computer readable medium of claim 12 , wherein the method further comprises:
receiving a user-defined item constraint; and
generating the recommendation based on the user-defined item constraint.

14. The non-transitory computer readable medium of claim 12 , wherein:
the community is split into a plurality of disjoint sampling groups;
each particular disjoint sampling group in the plurality of disjoint sampling groups has a particular size;
particular sizes decrease exponentially from a first size to a second size; and
the sampling group selected is a disjoint sampling group with a smallest size such that an accuracy level of the plurality of perturbed rating data is one of greater than and equal to a user-defined accuracy level.

15. The non-transitory computer readable medium of claim 12 , wherein the method further comprises:
receiving a user-defined recommendation algorithm; and
generating the recommendation based on the user-defined recommendation algorithm.

16. The non-transitory computer readable medium of claim 12 , wherein selecting a sampling group comprises:
splitting the community into a plurality of disjoint sampling groups, wherein each particular disjoint sampling group in the plurality of disjoint sampling groups has a particular size, and wherein particular sizes decrease exponentially from a largest size to a smallest size; and
selecting a particular disjoint sampling group based on a user-defined sampling rate.

Assignees

• Ramakrishnan Kadangode K
• Srivastava Divesh
• Cho Tae Won
• Zhang Yin
• At & T Ip I Lp
• Univ Texas

Inventors

• Ramakrishnan Kadangode K
• Srivastava Divesh
• Cho Tae Won
• Zhang Yin

Classifications

• G06N7/01 Primary

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06F16/244

  Grouping and aggregation · CPC title

• H04N21/4756

  for rating content, e.g. scoring a recommended movie · CPC title

• G11B27/105 Primary

  of operating discs · CPC title

• G06F16/48

  Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually · CPC title