Systems and methods for semantic inference and reasoning

• US 9,015,080 B2
• Filed: 03/16/2012
• Issued: 04/21/2015
• Est. Priority Date: 03/16/2012
• Status: Active Grant

• Current Assignee: Contiem Incorporated
• Original Assignee: Orbis Technology
• Inventors: Joshi, Sameer, Pehle, Todd, Crochet, Larry
• Primary Examiner(s): Gaffin, Jeffrey A
• Assistant Examiner(s): OLUDE AFOLABI, OLATOYOSI
• Application Number:
  US13/422,962
• Publication Number:
  US 20130246315A1
• Time in Patent Office: 1,131 Days
• Field of Search: 706/10, 706/55
• US Class Current: 706/10
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/04 Inference or reasoning models

Abstract

A method and system for analyzing a corpus of data artifacts is disclosed. The method comprises obtaining, by a computer, a semantic representation of the data artifacts, where the semantic representation indicates (1) entities identified in the data artifacts, and (2) semantic relationships among the entities as indicated by the data artifacts. The method further comprises clustering the data artifacts into clusters of semantically related data artifacts based on the semantic representation and inferring additional semantic relationships between pairs of the entities. The inferring comprises applying, on a cluster-by-cluster basis, a multi-tiered network of inference engines to a portion of the semantic representation corresponding to the cluster, where the multi-tiered network of inference engines includes a domain-independent inference tier and a domain-specific inference tier.

Claims

1. A method of analyzing a corpus of data artifacts, the method comprising:

obtaining, by a computer, a semantic representation of the data artifacts, wherein the semantic representation indicates;

entities identified in the data artifacts; and

semantic relationships among the entities, as indicated by the data artifacts;

clustering the data artifacts, by a computer, into clusters of semantically related data artifacts based on the semantic representation; and

inferring, by a computer, additional semantic relationships between pairs of the entities, wherein the inferring comprises;

on a cluster-by-cluster basis, applying a multi-tiered network of inference engines to a portion of the semantic representation corresponding to the cluster, wherein the multi-tiered network of inference engines comprises a domain-independent inference tier and a domain-specific inference tier,

wherein applying the multi-tiered network of inference engines comprises;

applying two or more inference engines sequentially, in parallel, or iteratively according to a static or dynamic schedule.

2. The method of claim 1, wherein obtaining the semantic representation comprises:

applying natural language processing techniques to extract the entities and relationships from natural language content contained of the data artifacts.

3. The method of claim 1, wherein obtaining the semantic representation comprises:

determining that the same entity is identified in the artifacts using different identifiers; and

disambiguating the entity by replacing one or more of the different identifiers with a common identifier for the entity.

4. The method of claim 1, wherein clustering the data artifacts comprises:

performing a semantic analysis to determine semantic interrelatedness of the data artifacts based on respective ones of the entities and relationships in the data artifacts.

5. The method of claim 1, wherein clustering the data artifacts comprises:

performing a syntactic analysis to determine syntactic interrelatedness of the data artifacts based on syntactic overlap of respective content of the data artifacts.

6. The method of claim 1, wherein the static or dynamic schedule is defined in one or more runtime configuration files.

7. The method of claim 1, wherein applying the multi-tiered network of inference engines comprises:

applying a plurality of domain-independent inference engines in the domain-independent tier and subsequently applying a plurality of inference engines in the domain-specific tier.

8. The method of claim 1, wherein the clustering and inferring are implemented using a parallel programming and execution model.

9. The method of claim 8, wherein the parallel programming and execution model corresponds to a MapReduce framework.

10. The method of claim 1, further comprising:

receiving, by a computer, a query identifying one or more of the entities;

identifying, by a computer in response to the query, a cluster of data artifacts most relevant to the one or more entities, wherein the identifying is dependent on one or more of the obtained or ;

inferred semantic relationships.

11. A system comprising:

a parallel processing facility comprising;

a plurality of computer processing cores; and

one or more memories coupled to the computer processing cores and storing program instructions executable by the processing cores to implement a semantic inference and reasoning engine configured to analyze a corpus of data artifacts by;

obtaining a semantic representation of the data artifacts, wherein the semantic representation indicates;

entities identified in the data artifacts; and

semantic relationships among the entities, as indicated by the data artifacts;

clustering the data artifacts into clusters of semantically related data artifacts based on the semantic representation; and

inferring additional semantic relationships between pairs of the entities, wherein the inferring comprises;

on a cluster-by-cluster basis, applying a multi-tiered network of inference engines to a portion of the semantic representation corresponding to the cluster, wherein the multi-tiered network of inference engines comprises a domain-independent inference tier and a domain-specific inference tier,

wherein applying the multi-tiered network of inference engines comprises applying two or more inference engines sequentially, in parallel, or iteratively according to a static or dynamic schedule.

12. The system of claim 11 further comprising a distributed storage facility coupled to the parallel processing facility and storing the corpus of data artifacts.

13. The system of claim 12, wherein the storage facility comprises a distributed file system.

14. The system of claim 11, wherein the parallel processing facility comprises at least one of:

a compute cluster, a superscalar supercomputer, a desktop grid, or a compute cloud.

15. The system of claim 11, wherein one or more memories further store program instructions executable to implement a parallel computation scheduling framework for executing the semantic inference and reasoning engine on the parallel processing facility using a MapReduce pattern.

16. The system of claim 11, wherein obtaining the semantic representation comprises:

applying natural language processing techniques to extract the entities and relationships from natural language content contained of the data artifacts.

17. The system of claim 11, wherein clustering the data artifacts comprises at least one of:

performing a semantic analysis to determine semantic interrelatedness of the data artifacts based on respective ones of the entities and relationships in the data artifacts; and

performing a syntactic analysis to determine syntactic interrelatedness of the data artifacts based on syntactic overlap of respective content of the data artifacts.

18. The system of claim 11, wherein applying the multi-tiered network of inference engines comprises:

applying a plurality of domain-independent inference engines in the domain-independent tier and subsequently applying a plurality of inference engines in the domain-specific tier.

19. A method of analyzing a corpus of data artifacts, the method comprising:

obtaining, by a computer, a semantic representation of the data artifacts, wherein the semantic representation indicates;

entities identified in the data artifacts; and

semantic relationships among the entities, as indicated by the data artifacts;

wherein obtaining the semantic representation comprises;

determining that the same entity is identified in the artifacts using different identifiers; and

disambiguating the entity by replacing one or more of the different identifiers with a common identifier for the entity;

clustering the data artifacts, by a computer, into clusters of semantically related data artifacts based on the semantic representation; and

inferring, by a computer, additional semantic relationships between pairs of the entities, wherein the inferring comprises;

on a cluster-by-cluster basis, applying a multi-tiered network of inference engines to a portion of the semantic representation corresponding to the cluster, wherein the multi-tiered network of inference engines comprises a domain-independent inference tier and a domain-specific inference tier.

20. A method of analyzing a corpus of data artifacts, the method comprising:

obtaining, by a computer, a semantic representation of the data artifacts, wherein the semantic representation indicates;

entities identified in the data artifacts; and

semantic relationships among the entities, as indicated by the data artifacts;

clustering the data artifacts, by a computer, into clusters of semantically related data artifacts based on the semantic representation, wherein clustering the data artifacts comprises at least one of;

performing a semantic analysis to determine semantic interrelatedness of the data artifacts based on respective ones of the entities and relationships in the data artifacts; and

performing a syntactic analysis to determine syntactic interrelatedness of the data artifacts based on syntactic overlap of respective content of the data artifacts; and

inferring, by a computer, additional semantic relationships between pairs of the entities, wherein the inferring comprises;

on a cluster-by-cluster basis, applying a multi-tiered network of inference engines to a portion of the semantic representation corresponding to the cluster, wherein the multi-tiered network of inference engines comprises a domain-independent inference tier and a domain-specific inference tier.

21. A method of analyzing a corpus of data artifacts, the method comprising:

obtaining, by a computer, a semantic representation of the data artifacts, wherein the semantic representation indicates;

entities identified in the data artifacts; and

semantic relationships among the entities, as indicated by the data artifacts;

clustering the data artifacts, by a computer, into clusters of semantically related data artifacts based on the semantic representation; and

inferring, by a computer, additional semantic relationships between pairs of the entities, wherein the inferring comprises;

on a cluster-by-cluster basis, applying a multi-tiered network of inference engines to a portion of the semantic representation corresponding to the cluster, wherein the multi-tiered network of inference engines comprises a domain-independent inference tier and a domain-specific inference tier,

wherein applying the multi-tiered network of inference engines comprises;

applying a plurality of domain-independent inference engines in the domain-independent tier and subsequently applying a plurality of inference engines in the domain-specific tier.