Search engine with augmented relevance ranking by community participation

1. A method of ranking one or more items in a result set, the method comprising:

receiving a query from a user comprising one or more search terms;

traversing an index in response to the query, the index comprising a location of each of a plurality of content items, words parsed from each of the plurality of content items, and at least one of tags and annotations regarding each of the plurality of content items, specified by a plurality of users, the at least one of the tags and annotations parsed from a plurality of user profiles;

identifying users in a social network of the user, the social network of the user determined by identifying at least one other user with whom the user has a direct relationship and a plurality other users with whom the user has an indirect relationship to a threshold number of degrees of separation, the social network identified by traversing a user profile data store comprising the plurality of user profiles, each of the plurality of user profiles comprising personalized interaction data regarding content items, including a location, tags, annotations, saved pages, and navigation history associated with each of a plurality of content items, and personalized social interaction data, including trust values associated with the other users with whom the user has a direct or indirect relationship;

generating a first result set responsive to the query, the first result set comprising a list of content items identified by traversing the index of content items, the first result set generated based on relevance of the one or more terms of the query to the words parsed from each of the plurality of content items, and the at least one of tags and annotations regarding each of the plurality of content items;

generating a second result set responsive to the query, the second result set comprising a list of content items identified by traversing the user profiles within the social network of the user, the second result set generated based on relevance of the one or more terms of the query with personalized information regarding content items in the user profiles within the social network of the user;

selecting an item from the second result set;

selecting a user profile from the plurality of user profiles, the user profile selected based on an indicia in the user profile of the item selected from the second result set;

selecting one or more items of personalized social interaction data and content item interaction data;

calculating a rank for the item in the second results set on the basis of at least on the content item interaction data and a trust value between the user submitting the query and the at least one other user associated with the selected user profile;

ranking the selected item in the second result set in accordance with the calculated rank; and

causing at least one interface screen to be displayed to the user comprising the list of content items in the first result set and the list of content items in the second result set.

2. The method of claim 1 wherein the social network of the user is determined by indirect relationships up to a threshold number of degrees of separation specified by the user.

3. The method of claim 1 wherein the index is an inverted index.

4. The method of claim 3 wherein said inverted index comprises aggregate personalized information from a plurality of users.

5. The method of claim 4 further comprising selecting at least one additional user profile, said additional user profile comprising one or more items of personalized information and calculating a rank for the items in the second result set on the basis of the first personalized data and the second personalized data.

6. The method of claim 1 wherein the trust value is computed by determining a feature vector for a plurality of users from information contained in the profile of each of the users, the plurality of users comprising the user of the selected profile and the user submitting the query, determining a similarity measure between the feature vectors of the user of the selected profile and the user submitting the query that an implied relationship exists between the user of the selected profile and the at least one other user, and the trust value is computed on the basis of the similarity measure between vectors.

7. The method of claim 1 wherein computing a trust value comprises utilizing a machine-learned trust function.

8. A system for ranking one or more items in a result set, the system comprising:

at least one search provider computer coupled to a plurality of client devices over a network, the search provider computer operative to:

receive a query from a user comprising one or more search terms;

traverse an index in response to the query, the index comprising a location of each of a plurality of content items, words parsed from each of the plurality of content items, and at least one of tags and annotations regarding each of the plurality of content items, specified by a plurality of users, the at least one of the tags and annotations parsed from a plurality of user profiles;

identify users in a social network of the user, the social network of the user determined by identifying at least one other user with whom the user has a direct relationship and a plurality other users with whom the user has an indirect relationship up to a threshold number of degrees of separation, the social network identified by traversing a user profile data store comprising the plurality of user profiles, each of the plurality of user profiles comprising personalized interaction data regarding content items, including a location, tags, annotations, saved pages, and navigation history associated with each of a plurality of content items, and personalized social interaction data, including trust values associated with the other users with whom the user has a direct or indirect relationship;

generate a first result set responsive to the query, the first result set comprising a list of content items identified by traversing the index of content items, the first result set generated based on relevance of the one or more terms of the query to the words parsed from each of the plurality of content items, and the at least one of tags and annotations regarding each of the plurality of content items;

generate a second result set responsive to the query, the second result set comprising a list of content items identified by traversing the user profiles within the social network of the user, the second result set generated based on relevance of the one or more terms of the query with personalized information regarding content items in the user profiles within the social network of the user;

select an item from the second result set;

select a user profile from the plurality of user profiles, the user profile selected based on an indicia in the user profile of the item selected from the second result set;

select one or more items of personalized social interaction data and content item interaction data;

calculate a rank for the item in the second results set on the basis of at least on the content item interaction data and a trust value between the user submitting the query and the at least one other user associated with the selected user profile;

rank the selected item in the second result set in accordance with the calculated rank; and

cause at least one interface screen to be displayed to the user comprising the list of content items in the first result set and the list of content items in the second result set.

9. The system of claim 8 wherein the social network of the user is determined by indirect relationships up to a threshold number of degrees of separation specified by the user.

10. The system of claim 8 wherein the index is an inverted index.

11. The system of claim 10 wherein said inverted index comprises aggregate personalized information from a plurality of users.

12. The system of claim 11 wherein said search provider is further operative to select at least one additional user profile, said additional user profile comprising one or more items of personalized information and calculating a rank for the items in the second result set on the basis of the first personalized data and the second personalized data.

13. The system of claim 8 wherein the trust value is computed by determining a feature vector for a plurality of users from information contained in the profile of each of the users, the plurality of users comprising the user of the selected profile and the user submitting the query, determining a similarity measure between the feature vectors of the user of the selected profile and the user submitting the query that an implied relationship exists between the user of the selected profile and the at least one other user, and the trust value is computed on the basis of the similarity measure between vectors.

14. The system of claim 13 wherein computing a trust value comprises utilizing a machine-learned trust function.