Method for content presentation

1. A method for personalizing an electronic program guide (EPG), the method comprising:

establishing a first matrix populated by ratings associated with a first plurality of v viewers in a content distribution system and a second plurality of p products each of which is a content item, the first matrix having dimensions v and p, each entry in the first matrix comprising either an empty entry or a rating indicating viewer interest in a content item;

bi-clustering the first matrix, thereby producing a second matrix, the second matrix comprising a matrix of ratings comprising a grouping of the viewers according to similarity of products chosen by the viewers and a grouping of the products according to similarity of viewers who viewed the products;

scoring metadata components associated with the products according to the matrix of ratings and producing a resulting score for each metadata component;

establishing, based at least in part on the score for each metadata component, a third matrix, the third matrix comprising a matrix of groups of viewers according to metadata scores;

establishing an auxiliary matrix of viewers according to viewer profile parameters, each entry in the auxiliary matrix indicating a viewer profile of an associated viewer and comprising a particular viewer profile parameter;

grouping together matrix entries of viewers with similar viewer profile parameters in the auxiliary matrix, thereby producing a fourth matrix, the fourth matrix comprising a matrix of groups of viewers according to viewer profile parameters;

for each of the products:

producing a product grid for each viewer group by sorting program groups for each viewer group, according to entries in the second matrix; and

identifying, for each group of viewers, a plurality of metadata parameters most likely to be associated with each viewer among the groups of viewers;

affiliating each individual viewer among the plurality of viewers with a viewer group;

obtaining a personal EPG view for at least one individual viewer, based, at least in part, on the affiliation of the individual viewer with a viewer group and the metadata parameters of the products; and

displaying the personal EPG view for the at least one individual viewer,

wherein the metadata parameters of the products comprise the identified most informative metadata parameters for a group of viewers to which the at least one viewer belongs.

2. The method according to claim 1 and wherein the rating comprises a measure of interest of at least one individual viewer in a product.

3. The method according to claim 2 and wherein the rating comprises a fraction, the fraction comprising a ratio of time the at least one individual viewer viewed the product to a total length of the product.

4. The method according to claim 1 and wherein the second matrix also comprises time-of-day information.

5. The method according to claim 1 and wherein the metadata components comprise:

title;

leading actor;

leading actress;

artist;

director;

viewing price;

length;

genre;

country of production; and

year of production.

6. The method according to claim 1 and wherein an unrated product is associated with rated products by determining which grouping of the products comprising rated products comprises metadata most closely matching metadata of the unrated product.

7. The method according to claim 6 and wherein the comparison of unrated product with rated products is performed giving different weights to different metadata parameters.

8. The method according to claim 7 and wherein the weighting of metadata parameters differs for different groups of viewers.

9. The method according to claim 8 and wherein the weight of metadata parameters is determined according to a statistical correlation of metadata parameter with program ratings within a given group of viewers.

10. The method according to claim 6 and wherein the unrated programs are associated with groups of viewers by cross-referencing the grouping of the products associated with the unrated product with groups of viewers in the second matrix, such that the associated groups of viewers comprise groups of viewers comprising high ratings for the grouping of the products associated with the unrated product.

11. The method according to claim 1 and wherein each viewer among the plurality of viewers is associated with one group of viewers by comparing ratings derived from a viewer associated viewing log with ratings of each viewer group, and associating each viewer with a group of viewers, such that the ratings of the associated group of viewers most closely resembles the ratings of the viewer.

12. A system for personalizing an electronic program guide (EPG), the system comprising:

a first matrix populating unit, operative to populate a first matrix by ratings associated with a first plurality of v viewers in a content distribution system and a second plurality of p products each of which is a content item, the first matrix having dimensions v and p, each entry in the first matrix comprising either an empty entry or a rating indicating viewer interest in a content item;

a bi-clustering unit, operative to bi-cluster the first matrix, thereby producing a second matrix, the second matrix comprising a matrix of ratings comprising a grouping of the viewers according to similarity of products chosen by the viewers and a grouping of the products according to similarity of viewers who viewed the products;

a metadata scoring apparatus operative to score metadata components associated with the products according to the matrix of ratings and produce a resulting score for each metadata component;

a third matrix establishing unit, operative to establish, based at least in part on the score for each metadata component, a third matrix, the third matrix comprising a matrix of groups of viewers according to metadata scores;

an auxiliary matrix establishing unit, operative to establish an auxiliary matrix of viewers according to viewer profile parameters, each entry in the auxiliary matrix indicating a viewer profile of an associated viewer and comprising a particular viewer profile parameter;

a matrix entry grouper operative to group together matrix entries of viewers with similar viewer profile parameters in the auxiliary matrix, thereby producing a fourth matrix, the fourth matrix comprising a matrix of groups of viewers according to viewer profile parameters;

a processor, which, for each of the products:

produces a product grid for each viewer group by sorting program groups for each viewer group, according to entries in the second matrix; and

identifies, for each group of viewers, a plurality of metadata parameters most likely to be associated with each viewer among the groups of viewers;

an affiliating unit which affiliates each individual viewer among the plurality of viewers with a viewer group;

a personal EPG view producer operative to produce a personal EPG view for at least one individual viewer, based, at least in part, on the affiliation of the individual viewer with a viewer group and the metadata parameters of the products; and

a display unit operative to display the personal EPG view for the at least one individual viewer,

wherein the metadata parameters of the products comprise the identified most informative metadata parameters for a group of viewers to which the at least one viewer belongs.

13. A system for personalizing an electronic program guide (EPG), the system comprising:

means for establishing a first matrix populated by ratings associated with a first plurality of v viewers in a content distribution system and a second plurality of p products each of which is a content item, the first matrix having dimensions v and p, each entry in the first matrix comprising either an empty entry or a rating indicating viewer interest in a content item;

means for bi-clustering the first matrix, thereby producing a second matrix, the second matrix comprising a matrix of ratings comprising a grouping of the viewers according to similarity of products chosen by the viewers and a grouping of the products according to similarity of viewers who viewed the products;

means for scoring metadata components associated with the products according to the matrix of ratings and producing a resulting score for each metadata component;

means for establishing, based at least in part on the score for each metadata component, a third matrix, the third matrix comprising a matrix of groups of viewers according to metadata scores;

means for establishing an auxiliary matrix of viewers according to viewer profile parameters, each entry in the auxiliary matrix indicating a viewer profile of an associated viewer and comprising a particular viewer profile parameter;

means for grouping together matrix entries of viewers with similar viewer profile parameters in the auxiliary matrix, thereby producing a fourth matrix, the fourth matrix comprising a matrix of groups of viewers according to viewer profile parameters;

means for processing each of the products, the means for processing comprising:

means for producing a product grid for each viewer group by sorting program groups for each viewer group, according to entries in the second matrix; and

means for identifying, for each group of viewers, a plurality of metadata parameters most likely to be associated with each viewer among the groups of viewers;

means for affiliating each individual viewer among the plurality of viewers with a viewer group;

means for obtaining a personal EPG view for at least one individual viewer, based, at least in part, on the affiliation of the individual viewer with a viewer group and the metadata parameters of the products; and

means for displaying the personal EPG view for the at least one individual viewer,

wherein the metadata parameters of the products comprise the identified most informative metadata parameters for a group of viewers to which the at least one viewer belongs.