System and Method for Identifying Groups of Entities
AssigneePeter MARBACH
InventorsPeter Marbach, Yueli Zhang
Life of this patent
Estimating the term…
Abstract
A computer implemented system and method is used to identify groups of entities based on bipartite directed graphs. Each group consists of two sets of entities, which have some association between each other. Two parameters, typicality and coverage, are taken into account in the computation for identifying the groups.
Claims20
A method for determining whether a group of entities can be extracted from an initial set of data, the initial set of data represented by a bipartite graph comprising a Set A and a Set B, the method performed by a computing device having access to associations between the entities of the Set A and the Set B, the method comprising:
- obtaining the initial set of data comprising a subset SA1 of the Set A, the subset SA1 comprising one or more entities;
- computing a process to identify other entities within the group comprising:
- computing a subset SB1 of the Set B that is associated with the subset SA1;
- computing a subset SA2 of the Set A that is associated with the subset SB1;
- and
- upon stopping the process, determining if SA2 and SB1 are within the group.
A method of computing a typicality value and a coverage value of an entity of a second set of entities, the method comprising:
- obtaining a first set of entities, a subset of the first set, and the entity of the second set, wherein the first set of entities and the second set of entities are represented in a bipartite directed graph;
- obtaining one or more associations between the entity of the second set and the first set;
- obtaining one or more associations between the entity of the second set and the subset of the first set;
- obtaining one or more associations between the subset of the first set and the second set;
- computing the typicality value by comparing:
- the associations between the entity of the second set and the subset of the first set with
- the associations between the entity of the second set and the first set;
- and,
- computing the coverage value by comparing:
- the associations between the entity of the second set and the subset of the first set with
- the associations between the subset of the first set and the second set.
Dependencies are inferred from the claim text.