Relationship discovery engine
AssigneeYahoo! Inc.
InventorsTed E. Dunning, Bradley D. Kindig
Life of this patent
Estimating the term…
Abstract
A system, method, and computer program product discover relationships among items and recommend items based on the discovered relationships. The recommendations provided by the present invention are based on user profiles that take into account actual preferences of users, without requiring users to complete questionnaires. An improved binomial log likelihood ratio analysis technique is applied, to reduce adverse effects of overstatement of coincidence and predominance of best sellers. The invention may be used, for example, to generate track lists for a personalized radio station.
Claims88
A computer-implemented method of generating a playlist, comprising:
- accepting user input specifying at least one track;
- selecting, from a corpus of tracks, a set of tracks responsive to relative degree of co-occurrence with the specified at least one track;
- arranging the set of tracks in a sequence;
- and
- outputting the sequence.
A computer-implemented method of generating a playlist, comprising:
- accepting user input specifying at least one artist;
- selecting, from a corpus of artists, a set of artists responsive to relative degree of co-occurrence with the specified at least one artist;
- and
- outputting the set of artists.
A computer-implemented method of generating a playlist, comprising:
- accepting user input specifying at least one artist;
- selecting, from a corpus of artists, a set of artists responsive to relative degree of co-occurrence with the specified at least one artist;
- selecting, for each of at least a subset of the artists, at least one track;
- arranging the selected tracks in a sequence;
- and
- outputting the sequence.
A computer-implemented method of generating a playlist, comprising:
- accepting user input specifying at least one track;
- determining at least one artist for the specified at least one track;
- selecting, from a corpus of artists, a set of artists responsive to relative degree of co-occurrence with the determined at least one artist;
- selecting, for each of at least a subset of the artists, at least one track;
- arranging the selected tracks in a sequence;
- and
- outputting the sequence.
A computer-implemented method of generating a playlist, comprising:
- accepting, from a user, a set of user preferences;
- selecting, from a corpus of tracks, a set of tracks responsive to the user preferences and responsive to descriptive information for the tracks;
- arranging the set of tracks in a sequence;
- and
- outputting the sequence.
Dependencies are inferred from the claim text.