System for browsing through a music catalog using correlation metrics of a knowledge base of mediasets
A system and method to navigate through a media item catalog and generate recommendations using behavioral metrics such as correlation metrics (FIGS. 1,2 ) from a knowledge base ( 400 ) of mediasets (FIG. 4, 1 - 7 ). One implementation uses a knowledge base comprising a collection of mediasets such as groupings of selected video items. Various metrics (Metric 1 -Metric m) among media items (m 1 . . . ) are considered by analyzing how the media items are grouped to form the sets in the knowledge base ( 400 ). Such metrics preferably are stored in a matrix ( 100 ) that allows the system to dynamically build appropriate navigation lists (FIG. 3 ) from media items that a user selects (FIG. 5 ).
1. A computer-implemented method of assisting a user in selecting items from an electronic catalog of video items, the method comprising:
(a) obtaining access to metric values derived from a knowledge base of predetermined mediasets associated with the electronic catalog; wherein the metric values reflect a level of association for selected pairs of video items within the knowledge base of mediasets, and wherein the metric values are not affected by any metadata descriptive of the said video items' content, but rather reflect a relative frequency with which the video items in the pair are associated together by various users;
(b) receiving an initial selection of at least one video item to define an initial input video set;
(c) generating an output video item navigation list responsive to at least one item of the input video set, based on the metric values derived from the knowledge base; and
(d) communicating the generated navigation list to a user.
2. The method of claim 1 further comprising the steps of:
(e) receiving from the user an instruction to edit the navigation list;
(f) editing the navigation list as instructed by the user to define a new input video set;
(g) responsive to the new input video set, generating a new navigation list based on the metric values; and
(h) communicating the new navigation list to the user.
3. The method according to claim 2 wherein said editing step includes adding a user-selected video item to the initial input video set to define the new input video set.
4. The method according to claim 2 wherein said editing step includes:
receiving from the user a selection of at least one video item on the navigation list for deletion; and
deleting the user-selected video item from the initial input video set to form the new input video set.
5. The method according to claim 1 wherein said communicating step comprises communication over a network for displaying the navigation list on a remote client device.
6. The method according to claim 1 wherein said communicating step comprises communication over a network for displaying the navigation list on a mobile client device.
7. The method according to claim 1 and further comprising ordering the generated navigation list according to the association metric values used to generate the navigation list.
8. The method according to claim 1 and further comprising filtering the generated navigation list responsive to predetermined constraints.
9. The method according to claim 1 wherein the initial input video set comprises a playlist.
10. The method according to claim 1 wherein the user comprises a remote software process rather than a person.
11. The method according to claim 2 including repeating steps (e) through (h) to realize an interactive browsing session.
12. The method according to claim 1 wherein the catalog of video items comprises html links to the video items.
13. The method according to claim 12 wherein the association metric values comprise co-concurrency metric values.
14. The method according to claim 12 wherein the association metric values comprise at least one of a pre-concurrency metric, a post-concurrency metric, and a combination of the foregoing metrics applied transitively.
15. The method according to claim 12 wherein the association metric values comprise co-concurrency metric values, the co-concurrency metric values indicating, for each pair of video items (i,j), how many times item i and item j appear together within any of the mediasets of the knowledge base.
16. A non-volatile, non-transient, computer-readable storage medium having computer-readable program code stored therein, said computer-readable program code including:
computer-readable program code for receiving a selection of a catalog for user browsing and for accessing a collection of metric data associated with the selected catalog;
wherein the metric data comprises metric values for pairs of media items derived from a knowledge base of mediasets;
wherein the metric values reflect a level of co-concurrency for each of the pairs of media items within the knowledge base of mediasets; and further wherein the co-concurrency metric values indicate, for each pair of media items (i,j), how many times item i and item j appear together within any of the mediasets of the knowledge base, so that the co-concurrency metric values are not affected by any descriptor, characteristic, classification or attribute of the media items' content;
computer-readable program code for receiving from the user an initial selection of at least one media item from the selected catalog to define an initial input media set;
computer-readable program code for generating a navigation list based on selected metrics for each item of the initial input media set; and
computer-readable program code for communicating the generated navigation list to the user.