Method and apparatuses for enabling recommendations

Patent metadata

Field	Value
Publication number	US-9594758-B2
Application number	US-201114349434-A
Country	US
Kind code	B2
Filing date	Oct 5, 2011
Priority date	Oct 5, 2011
Publication date	Mar 14, 2017
Grant date	Mar 14, 2017

Abstract

Method and a recommendation system ( 200 ) for making a media item recommendation to a user. A weighted graph is created ( 2:2 ) that comprises item nodes representing media items and edges linking the item nodes. Weights are also assigned to the edges based on weights of contributors shared by the linked item nodes. When an initial media item of interest to the user is identified ( 2:4 ), the recommendation system determines ( 2:5 ) a set of media items which are linked to the initial media item in a weighted graph. At least one of the determined set of media items is also selected based on the assigned weight of edges linking the media items to the initial media item. The recommendation system then provides ( 2:6 ) the media item recommendation with the at least one selected media item to the user.

First claim

The invention claimed is:

1. A method in a network node for making a media item recommendation to a user in a communication network, the method comprising:
identifying an initial media item deemed to be of interest to the user, wherein said initial media item is identified when given explicitly in a received recommendation request at least from a user terminal, or by detecting that it has been previously consumed and/or rated highly by the user;
determining a set of media items that are linked to the initial media item in a weighted graph, said weighted graph comprising item nodes representing media items and edges linking the item nodes, wherein weights are assigned to said edges based on weights of contributors shared by the linked item nodes;
selecting at least one of the determined set of media items based on the assigned weight of edges linking said set of media items to the initial media item;
providing the at least one selected media item in the media item recommendation to the user, wherein the weighted graph is created by:
creating a base graph comprising said item nodes and contributor nodes representing said contributors and being linked to the item nodes;
calculating weights for said item nodes and contributor nodes based on meta data relating to the corresponding media items and contributors;
replacing said contributor nodes with said edges linking the item nodes; and
creating the weighted graph by assigning weights to the edges based on the weights calculated for the replaced contributor nodes.

2. The method of claim 1 , wherein at least one media item having the highest weight of its edge linking the initial media item is selected from the determined set of media items for said recommendation.

3. The method of claim 2 , wherein each of a plurality of media items is ranked in the recommendation according to the weight of their edges linking the initial media item.

4. The method of claim 1 , wherein the weights are calculated for the item nodes and contributor nodes by using at least one of: Eigenvector Centrality and PageRank.

5. The method of claim 1 , wherein said meta data is obtained from one or more content providers providing the media items.

6. The method of claim 1 , wherein said weights are assigned to the edges further based on weights of awards mutual to the linked item nodes.

7. A recommendation system configured to make media item recommendations to users in a communication network, the recommendation system comprising:
one or more processors and a memory coupled to at least one of the one or more processors configured to:
create a weighted graph comprising item nodes and edges linking the item nodes, wherein weights are assigned to said edges based on weights of contributors shared by the linked item nodes;
identify an initial media item deemed to be of interest to a user, and to identify said initial media item when given explicitly in a received recommendation request at least from a user terminal, or by detecting that the initial media item has been previously consumed and/or rated highly by the user;
determine a set of media items which are linked to the initial media item in the weighted graph, and to select at least one of the determined set of media items based on the assigned weight of edges linking said set of media items to the initial media item; and
provide the at least one selected media item in the media item recommendation to the user, wherein the one or more processors are further configured to:
create a base graph comprising said item nodes and contributor nodes representing said contributors and being linked to the item nodes;
calculate weights for said item nodes and contributor nodes based on meta data relating to the corresponding media items and contributors;
replace said contributor nodes with said edges linking the item nodes; and
create the weighted graph by assigning weights to the edges based on the weights calculated for the replaced contributor nodes.

8. The recommendation system of claim 7 , wherein the one or more processors are further configured to select at least one media item having the highest weight of its edge linking the initial media item from the determined set of media items for said recommendation.

9. The recommendation system of claim 8 , wherein the one or more processors are further configured to rank a plurality of media items in the recommendation according to the weight of their edges linking the initial media item.

10. The recommendation system of claim 7 , wherein the one or more processors are further configured to calculate the weights for the item nodes and contributor nodes by using at least one of: Eigenvector Centrality and PageRank.

11. The recommendation system of claim 7 , wherein the one or more processors are further configured to obtain said meta data from one or more content providers providing the media items.

12. The recommendation system of claim 7 , wherein the one or more processors are further configured to assign said weights to the edges further based on weights of awards mutual to the linked item nodes.

Assignees

• Bararsani Azadeh
• Lidström Mattias
• Söderberg Joakim
• Ericsson Telefon Ab L M

Inventors

• Bararsani Azadeh
• Lidström Mattias
• Söderberg Joakim

Classifications

• G06F17/30029

  Physics · mapped topic

• G06F17/30017 Primary

  Physics · mapped topic

• G06F17/30958

  Physics · mapped topic

• G06F17/30053

  Physics · mapped topic

• G06F16/435

  Filtering based on additional data, e.g. user or group profiles · CPC title