Mixed-model recommender for leisure activities
One embodiment of the present invention provides a method for recommending leisure activities to a user. During operation, the system receives at least one query for leisure activities. The system then determines a collaborative filtering score of a candidate activity based on a collaborative filtering model, a soft query score for the candidate activity based on a soft query model, a content preference score for the candidate activity based on a content preference model and the user's past behavior, and a distance score for the candidate activity based on a distance model. Next, the system generates a composite score for the candidate activity by calculating a weighted average of the collaborative filtering score, the soft query score, the content preference score, and the distance score. The system further returns a recommendation list containing the activities with the highest composite scores.
1. A computer-implemented method for recommending leisure activities to a user, the method comprising:
receiving one query for leisure activities, which are associated with a plurality of activity types;
determining by a computer a collaborative rating of a candidate activity by other users;
determining by the computer the user's preference for the candidate activity based on a soft query on a set of attributes associated with the candidate activity;
determining by the computer the user's interest in the candidate activity based on the user's past behavior;
determining by the computer proximity of the candidate activity to the user's current location;
generating by the computer a composite score for the candidate activity based on the type of the candidate activity and a combination of the collaborative rating, the user's preference, the user's interest, and the proximity of the candidate activity; and
recommending to the user a list of leisure activities of different activity types based on composite scores of the leisure activities.
2. The method of claim 1 , wherein determining the collaborative rating comprises receiving a set of user-profiling information.
3. The method of claim 1 , wherein determining the collaborative rating comprises retrieving one or more ratings for the candidate activity submitted by other users in a same profile group to which the user is matched.
4. The method of claim 1 , wherein determining the user's preference comprises receiving preference information from the user.
5. The method of claim 1 , wherein determining the user's interest comprises:
extracting a set of keywords from a set of content previously accessed by the user; and
determining whether a description of the candidate activity matches any of the keywords.
6. The method of claim 1 , wherein determining the proximity comprises computing a distance between a location associated with the candidate activity and a location associated with the user.
7. The method of claim 6 , wherein determining the proximity further comprises:
receiving a set of GPS coordinates; and
computing a motion range for the user.
8. A computer system for recommending leisure activities to a user, the computer system comprising:
a processor;
a memory;
a receiving mechanism configured to receive at least one query for leisure activities, which are associated with a plurality of activity types;
a collaborative rating mechanism configured to determine a collaborative rating of a candidate activity by other users;
a preference determination mechanism configured to determine the user's preference for the candidate activity based on a soft query on a set of attributes associated with the candidate activity;
an interest determination mechanism configured to determine the user's interest for the candidate activity based on the user's past behavior;
a proximity determination mechanism configured to determine proximity of the candidate activity to the user's current location;
a composite scoring mechanism configured to generate a composite score for the candidate activity based on the type of the candidate activity and a combination of the collaborative rating, the user's preference, the user's interest, and the proximity of the candidate activity; and
a recommendation mechanism configured to recommend a list of leisure activities of different activity types based on composite scores of the leisure activities.
9. The computer system of claim 8 , wherein while determining the collaborative rating, the collaborative rating mechanism is further configured to receive a set of user-profiling information.
10. The computer system of claim 8 , wherein while determining the collaborative rating, the collaborative rating mechanism is further configured to retrieve one or more ratings for the candidate activity submitted by other users in a same profile group to which the user is matched.
11. The computer system of claim 8 , wherein while determining the user's preference, the preference determination mechanism is configured to receive preference information from the user.
12. The computer system of claim 8 , wherein while determining the user's interest, the interest determination mechanism is further configured to:
extract a set of keywords from a set of content previously accessed by the user; and
determine whether a description of the candidate activity matches any of the keywords.
13. The computer system of claim 8 , wherein while determining the proximity, the proximity determination mechanism is further configured to compute a distance between a location associated with the candidate activity and a location associated with the user.
14. The computer system of claim 13 , wherein while determining the proximity, the proximity determination mechanism is further configured to:
receive a set of GPS coordinates; and
compute a motion range for the user.
15. A computer readable storage medium storing instructions which when executed by a computer cause the computer to perform a method for recommending leisure activities to a user, the method comprising:
receiving one query for leisure activities, which are associated with a plurality of activity types;
determining by a computer a collaborative rating of a candidate activity by other users;
determining by the computer the user's preference for the candidate activity based on a soft query on a set of attributes associated with the candidate activity;
determining by the computer the user's interest in the candidate activity based on the user's past behavior;
determining by the computer proximity of the candidate activity to the user's current location;
generating by the computer a composite score for the candidate activity based on the type of the candidate activity and a combination of the collaborative rating, the user's preference, the user's interest, and the proximity of the candidate activity; and
recommending to the user a list of leisure activities of different activity types based on composite scores of the leisure activities.
16. The computer readable storage medium of claim 15 , wherein determining the collaborative rating comprises receiving a set of user-profiling information.
17. The computer readable storage medium of claim 15 , wherein determining the collaborative rating comprises retrieving one or more ratings for the candidate activity submitted by other users in a same profile group to which the user is matched.
18. The computer readable storage medium of claim 15 , wherein determining the user's preference comprises receiving preference information from the user.
19. The computer readable storage medium of claim 15 , wherein determining the user's interest comprises:
extracting a set of keywords from a set of content previously accessed by the user; and
determining whether a description of the candidate activity matches any of the keywords.
20. The computer readable storage medium of claim 15 , wherein determining the proximity comprises computing a distance between a location associated with the candidate activity and a location associated with the user.
21. The computer readable storage medium of claim 20 , wherein determining the proximity further comprises:
receiving a set of GPS coordinates; and
computing a motion range for the user.