Context aware recommendation

1. A computer system including instructions recorded on a non-transitory computer-readable medium and executable by at least one processor, the system comprising:

a context aware recommendation manager configured to cause the at least one processor to provide one or more recommendations to a user in response to a query related to the user by integrating contextual information of a context of the user in a recommendation model while considering a granular structure of the context and the contextual information thereof, structural elements of the granular structure including at least a multiple of granular elements for location arranged in a hierarchy of levels along a location context granularity path, and a multiple of granular elements for time arranged in a hierarchy of levels along a time context granularity path, wherein context combination paths of the context are formed by a cross product of all context granularity paths in the granular structure, and wherein the context aware recommendation manager includes:

a multi-granular context module configured to detect the context of the user related to the query and determine the granular structure of the context including one or more structural elements of the granular structure based on the contextual information that characterizes the information related to the user;

a pre-filtering recommendation paradigm module configured to pre-filter the contextual information according one or more context combination paths between different structural elements of the granular structure and select historical network navigation data related to the user for comparison of the context combination paths; and

a recommendation builder configured to:

model each context combination path for the query in the recommendation model by calculating a recommendation performance for each different context combination path as compared with the selected historical network navigation data while considering the granular structure of the context and the contextual information thereof, and

determine at least one of the context combination paths as having a best recommendation performance to provide the one or more recommendations to the user in response to the query.

2. The system of claim 1 , wherein the context includes multiple contexts and each of the multiple contexts includes a hierarchical structure with a multi-granular organization.

3. The system of claim 1 , wherein the contextual information associated with the context reveals an intent of the user for a query, the contextual information including time and location of the query issued by the user, and the recommendation builder is configured to select the best context combination for <query, location, time>.

4. The system of claim 1 , wherein the contextual information associated with the context includes a profile of the user including one or more of gender of the user, age of the user, personal interest of the user, salary of the user, likes and dislikes of the user, and one or more social networks associated with the user.

5. The system of claim 1 , wherein the structural elements of the granular structure of the context and the contextual information related to the user include one or more granular levels and one or more granular components within each granular level.

6. The system of claim 1 , wherein the multi-granular context module is configured to automatically detect the context related to the user based on the historical network navigation data related to the user.

7. The system of claim 1 , wherein the multi-granular context module is configured to automatically detect dependencies between the structural elements of the context and the contextual information related to the user including dependencies between one or more granular levels and one or more granular components within each granular level.

8. The system of claim 1 , the recommendation builder further comprising:

a trainer module configured to automatically discover one or more potentially predictive relationships and dependencies between the structural elements of the granular structure of the context and the contextual information related to the user; and

a recommender module configured to automatically detect the one or more best context combinations for the query as a context aware recommendation provided to the user.

9. A computer-implemented method, comprising:

providing one or more recommendations to a user in response to a query related to the user by integrating contextual information of a context related to the user in a recommendation model while considering a granular structure of the context and the contextual information thereof, structural elements of the granular structure including at least a multiple of granular elements for location arranged in a hierarchy of levels along a location context granularity path, and a multiple of granular elements for time arranged in a hierarchy of levels along a time context granularity path, wherein context combination paths of the context are formed by a cross product of all context granularity paths in the granular structure, and wherein the providing includes:

detecting the context of the user,

determining the granular structure of the context including one or more structural elements of the granular structure based on the contextual information that characterizes the information related to the user,

pre-filtering the contextual information according one or more context combination paths between different structural element of the granular structure,

selecting historical network navigation data related to the user for comparison with each of the context combination paths,

modeling each context combination path for the query in the recommendation model by calculating a recommendation performance for each different context combination path as compared with the selected historical network navigation data while considering the granular structure of the context and the contextual information thereof, and

determining at least one of the context combination paths as having a best recommendation performance to provide the one or more recommendations to the user in response to the query.

10. The method of claim 9 , wherein the context includes multiple contexts and each of the multiple contexts includes a hierarchical structure with a multi-granular organization.

11. The method of claim 9 , wherein the contextual information associated with the context reveals an intent of the user for a query, the contextual information including time and location of the query issued by the user, and the recommendation builder is configured to select the best context combination for <query, location, time>.

12. The method of claim 9 , wherein the contextual information associated with the context includes a profile of the user including one or more of gender of the user, age of the user, personal interest of the user, salary of the user, likes and dislikes of the user, and one or more social networks associated with the user.

13. The method of claim 9 , wherein the structural elements of the granular structure of the context and the contextual information related to the user include one or more granular levels and one or more granular components within each granular level.

14. The method of claim 9 , the method further comprising:

automatically detecting the context related to the user based on the historical network navigation data related to the user;

automatically detecting dependencies between the structural elements of the context and the contextual information related to the user including dependencies between one or more granular levels and one or more granular components within each granular level;

automatically discovering one or more potentially predictive relationships and dependencies between the structural elements of the granular structure of the context and the contextual information related to the user; and

automatically detecting the one or more best context combinations for the query as a context aware recommendation provided to the user.

15. A computer program product, the computer program product being tangibly embodied on a non-transitory computer-readable storage medium and including instructions that, when executed by at least one processor, are configured to:

provide one or more recommendations to a user in response to a query related to the user by integrating contextual information of a context related to the user in a recommendation model while considering a granular structure of the context and the contextual information thereof, structural elements of the granular structure including at least a multiple of granular elements for location arranged in a hierarchy of levels along a location context granularity path, and a multiple of granular elements for time arranged in a hierarchy of levels along a time context granularity path, wherein context combination paths of the context are formed by a cross product of all context granularity paths in the granular structure, by

detecting the context of the user related to the query,

determining the granular structure of the context including one or more structural elements of the granular structure based on the contextual information that characterizes the information related to the user,

pre-filtering the contextual information according one or more context combination paths between each structural element of the granular structure,

selecting historical network navigation data related to the user for comparison with each of the context combination paths,

modeling each context combination path for the query in the recommendation model by calculating a recommendation performance for each different context combination path as compared with the selected historical network navigation data while considering the granular structure of the context and the contextual information thereof, and

determining at least one of the context combination paths as having a best recommendation performance to provide the one or more recommendations to the user in response to the query.

16. The computer program product of claim 15 , wherein the context includes multiple contexts and each of the multiple contexts includes a hierarchical structure with a multi-granular organization.

17. The computer program product of claim 15 , wherein the contextual information associated with the context reveals an intent of the user for a query, the contextual information including time and location of the query issued by the user, and the recommendation builder is configured to select the best context combination for <query, location, time>.

18. The computer program product of claim 15 , wherein the contextual information associated with the context includes a profile of the user including one or more of gender of the user, age of the user, personal interest of the user, salary of the user, likes and dislikes of the user, and one or more social networks associated with the user.

19. The computer program product of claim 15 , wherein the structural elements of the granular structure of the context and the contextual information related to the user include one or more granular levels and one or more granular components within each granular level.

20. The computer program product of claim 15 , further comprising instructions that, when executed by the processor, are configured to:

automatically detect the context related to the user based on the historical network navigation data related to the user;

automatically detect dependencies between the structural elements of the context and the contextual information related to the user including dependencies between one or more granular levels and one or more granular components within each granular level;

automatically discover one or more potentially predictive relationships and dependencies between the structural elements of the granular structure of the context and the contextual information related to the user; and

automatically detect the one or more best context combinations for the query as a context aware recommendation provided to the user.