US 2012/0233180 A1 · SELECTING A DATA ELEMENT IN A NETWORK

Abstract

One embodiment of the disclosure can represent within a K-partite graph, weighting factors between a set of identifier elements and a set of data elements. The K-partite graph can also represent weighting factors between the set of identifier elements and a set of metadata elements. In the K-partite graph, the set of identifier elements, the set of data elements, and the set of metadata elements are independent and disjoint sets such that no two vertices within a same set are adjacent. A score vector can be calculated that ranks each data element in a set of data elements. The score vector can be calculated from the weighting factors. At least one data element from the set of data elements can be selected using the score vector and a predetermined selection criterion.

Claims22

Dependent claims

A method comprising:

representing within a K-partite graph, weighting factors between a set of identifier elements and a set of data elements, wherein the K-partite graph also represents weighting factors between the set of identifier elements and a set of metadata elements, wherein the K-partite graph the set of identifier elements, the set of data elements, and the set of metadata elements are independent and disjoint sets such that no two vertices within a same set are adjacent;
calculating a score vector that ranks each data element in a set of data elements, said score vector being calculated from the weighting factors;
and
selecting at least one data element from the set of data elements using the score vector and a predetermined selection criterion.

1 more element

A method comprising:

generating a K-partite graph to represent weighing factors between a set of identifier elements and a set of data elements, wherein said K-partite graph represents weighting factors between the set of identifier elements and a set of metadata elements, wherein each data element in the set of data elements is selected from a group consisting of:
(i) a multimedia clip;
(ii) a keyword;
(iii) a bookmark tag;
(iv) a text segment;
(v) a book recommendation;
and (vi) a product recommendation, and wherein each identifier element in the set of identifier elements is selected from a group consisting of:
(i) a user account;
(ii) a user profile;
(iii) a customer account;
(iv) a social networking profile;
(v) a medical health record;
(vi) a book;
and (vii) a journal article;
calculating a score vector that ranks each data element in the set of data elements, wherein the score vector is calculated from the weighting factors;
and
selecting at least one data element using the score vector and user-provided selection criteria.

14 more elements

Dependencies are inferred from the claim text.

Selecting a data element in a network

Life of this patent

Selecting a data element in a network

Life of this patent

Abstract

Claims22