Entity resolution
A relationship graph system disclosed herein provides a system and method for inferring relationships between various entities based on analysis of a plurality of observations including elements related to two different entities. The method comprises representing each of the plurality of observations using a relationship edge between two nodes, each of the two nodes representing an element related to one of the two different entities, assigning a score to each relationship edge based on metadata related to the observation represented by the relationship edge, combining a plurality of relationship edges between the elements related to the two different entities to generate an aggregate relationship edge between the two different entities, and inferring a relationship between the two different entities based on length of the aggregate relationship edge between the two different entities.
1. A computer implemented method comprising:
detecting, by one or more hardware processors of an information cataloging system, a plurality of observations including various elements related to different entities;
representing the one or more of various elements as nodes in a graph in the information cataloging system executed by one or more hardware processors by storing the observations in a database;
representing, using the one or more hardware processors of the information cataloging system, relations between one or more nodes as relation edges in the graph;
assigning, using the one or more hardware processors of the information cataloging system, confidence levels to one or more of the plurality of nodes;
assigning, using the one or more hardware processors of the information cataloging system, a score to each relationship edge based on metadata related to the relations represented by the relationship edge;
computing, using the one or more hardware processors of the information cataloging system, confidence values for one or more of the relationship edges based on confidence levels assigned to the nodes associated with the edges and the score assigned to each of the relationship edges;
traversing the graph using the one or more hardware processors of the information cataloging system to identify a related plurality of the nodes and a related plurality of relationship edges of each of the one or more entities at varying levels of confidence; and
cataloging the related plurality of the nodes and the related plurality of relationship edges for each of the one or more entities at the varying levels of confidence in the information cataloging system by storing the relationship in the database, wherein the information cataloging system is configured to respond to a user query based on the cataloged related plurality of the nodes and the cataloged related plurality of relationship edges for each of the one or more entities at the varying levels of confidence.
2. The method of claim 1 , further comprising assigning an identifier to each of the one or more entities, wherein an identifier is a randomly assigned value.
3. The method of claim 1 , further comprising assigning an identifier to each of the one or more entities, wherein an identifier is derived from one or more nodes in the graph.
4. The method of claim 3 , wherein the identifiers are represented as identifier nodes in the graph.
5. The method of claim 4 , further comprising computing higher level edges between identifier nodes and one or more other nodes.
6. The method of claim 5 , further comprising computing higher order edges between one or more pairs of nodes based on the existing one or more edges between the pairs of nodes.
7. The method of claim 6 , further comprising computing a confidence level of the computed higher order edges based on distance of the computed higher order edges.
8. The method of claim 4 , further comprising receiving a query to the graph to receive various relations for one or more of the entities.
9. The method of claim 8 , wherein the query specifies a confidence level for inferring relations for one or more of the entities.
10. The method of claim 9 , wherein the query results include associated confidence levels for the relations for the one of the entities.
11. The method of claim 10 , further comprising retrieving identifiers assigned to one or more of the query results.
12. The method of claim 8 , wherein the query is stored as new observation.
13. The method of claim 8 , wherein the query specifying multiple fields and confidence level associated with each of the multiple fields.
14. The method of claim 8 , wherein the query is added as a new observation in real-time or in a batch mode.
15. The method of claim 1 , wherein each of the entities represent at least one of an individual, a household, a company, a shared interest, and an organization.
16. A physical article of manufacture including one or more tangible computer-readable storage media, encoding computer-executable instructions for executing a computer process on a computer system comprising one or more hardware processors, the computer process comprising:
detecting a plurality of observations including various elements related to different entities;
representing the one or more of various elements as nodes in a graph in an information cataloging system executed by one or more hardware processors by storing the observations in a database;
representing relations between one or more nodes as relation edges in the graph;
assigning confidence levels to one or more of the plurality of nodes;
assigning, using the information cataloging system, a score to each relationship edge based on metadata related to the relations represented by the relationship edge;
computing confidence values for one or more of the relationship edges based on confidence levels assigned to the nodes associated with the edges and the score assigned to each of the relationship edges;
traversing the graph to identify a related plurality of the nodes and a related plurality of relationship edges of each of the one or more entities at varying levels of confidence; and
cataloging the related plurality of the nodes and the related plurality of relationship edges for each of the one or more entities at the varying levels of confidence in the information cataloging system by storing the relationship in the database, wherein the information cataloging system is configured to respond to a user query based on the cataloged related plurality of the nodes and the cataloged related plurality of relationship edges for each of the one or more entities at the varying levels of confidence.
17. The physical article of manufacture of claim 16 , wherein the computer process further comprising assigning an identifier to each of the one or more entities, wherein an identifier is a randomly assigned value.
18. The physical article of manufacture of claim 16 , wherein the computer process further comprising assigning an identifier to each of the one or more entities, wherein an identifier is derived from one or more nodes in the graph.
19. The physical article of manufacture of claim 16 , wherein the identifiers are represented as identifier nodes in the graph.
20. The physical article of manufacture of claim 16 , wherein the computer process further comprising computing higher level edges between identifier nodes and one or more other nodes.