Methods and systems for modeling complex taxonomies with natural language understanding

• US 9,495,345 B2
• Filed: 12/09/2015
• Issued: 11/15/2016
• Est. Priority Date: 12/09/2014
• Status: Active Grant

• Current Assignee: AI IP Investments Limited
• Original Assignee: Idibon, Inc.
• Inventors: Munro, Robert J., Erle, Schuyler D., Brenier, Jason, Long, Jessica D., Callahan, Brendan D., Tepper, Paul A., Nunez, Edgar, Schnoebelen, Tyler J.
• Primary Examiner(s): Guerra-Erazo, Edgar
• Application Number:
  US14/964,511
• Publication Number:
  US 20160162476A1
• Time in Patent Office: 342 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/243 Natural language query formulation
• CPC Class Codes: G06F 16/24532 of parallel queries
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/288 Entity relationship models
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 16/367 Ontology
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 3/0482 Interaction with lists of selectable items, e.g. menus
• CPC Class Codes: G06F 40/137 Hierarchical processing, e.g. outlines
• CPC Class Codes: G06F 40/169 Annotation, e.g. comment data or footnotes
• CPC Class Codes: G06F 40/221 Parsing markup language streams streaming G06F40/149
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30
• CPC Class Codes: G06F 40/42 Data-driven translation
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

Systems and methods are presented for the automatic placement of rules applied to topics in a logical hierarchy when conducting natural language processing. In some embodiments, a method includes: accessing, at a child node in a logical hierarchy, at least one rule associated with the child node; identifying a percolation criterion associated with a parent node to the child node, said percolation criterion indicating that the at least one rule associated with the child node is to be associated also with the parent node; associating the at least one rule with the parent node such that the at least one rule defines a second factor for determining whether the document is to also be classified into the parent node; accessing the document for natural language processing; and determining whether the document is to be classified into the parent node or the child node based on the at least one rule.

Claims

1. A method for improving natural language processing conducted by a natural language model, the method comprising:

accessing, at a first node in a logical hierarchy configured to guide classification of a plurality of documents by the natural language model, at least one rule associated with the first node, said at least one rule defining a first factor for determining whether a document among the plurality of documents is to be classified into the first node;

identifying a percolation criterion associated with a second node in the logical hierarchy that is a parent node to the first node, said percolation criterion indicating that the at least one rule associated with the first node is to be associated also with the second node;

based on the identified percolation criterion, associating the at least one rule with the second node such that the at least one rule defines a second factor for determining whether the document is to also be classified into the second node;

accessing the document for natural language processing; and

classifying the document according to the logical hierarchy by determining whether the document is to be classified into at least one of the second node and the first node based on the at least one rule associated with both the first node and the second node.

2. The method of claim 1, wherein the percolation criterion is a first percolation criterion, and the method further comprises:

identifying a second percolation criterion associated with a third node in the logical hierarchy that is a parent node to the second node, said percolation criterion indicating that the at least one rule associated with the first node is to be associated also with the third node; and

based on the identified second percolation criterion, associating the at least one rule with the third node such that the at least one rule defines a third factor for determining whether the document is to also be classified into the third node.

3. The method of claim 1, further comprising:

determining that a second percolation criterion is not associated with a third node in the logical hierarchy that is a parent node to the second node; and

based on said determination, limiting the at least one rule to be associated only to the first node and the second node.

4. The method of claim 1, wherein the percolation criterion includes the second node being repeated at least once within the logical hierarchy.

5. The method of claim 1, wherein the percolation criterion includes a metadata tag that is enabled.

6. The method of claim 1, wherein classifying the document according to the logical hierarchy comprises analyzing the document utilizing the at least one rule to determine if the document is to be classified into the second node.

7. The method of claim 6, wherein classifying the document according to the logical hierarchy further comprises:

determining that the document is to be classified into the second node based at least in part on the at least one rule; and

based on said determination, analyzing the document utilizing the at least one rule to determine if the document is to be classified into one or more child nodes of the second node, the one or more child nodes including the first node.

8. The method of claim 7, wherein determining that the document is to be classified into the second node based at least in part on the at least one rule further comprises:

determining at least one region of text within the document that satisfies the first factor specified by the at least one rule; and

temporarily storing the at least one region of text in a cache in response to the determination.

9. The method of claim 1, further comprising:

receiving an input to identify which documents in the plurality of documents are to be classified into the first node;

in response to said input, analyzing the plurality of documents by the natural language model, using the at least one rule, to determine which documents are to be classified into the first node; and

causing display of the documents that are determined to be classified into the first node.

10. The method of claim 9, further comprising causing display of documents that are determined to be classified into the first node, wherein the determination is made at least in part by the document triggering the at least one rule.

11. A system comprising:

a memory; and

a processor coupled to the memory and configured to;

access, at a first node in a logical hierarchy configured to guide classification of a plurality of documents by the natural language model, at least one rule associated with the first node, said at least one rule defining a first factor for determining whether a document among the plurality of documents is to be classified into the first node;

identify a percolation criterion associated with a second node in the logical hierarchy that is a parent node to the first node, said percolation criterion indicating that the at least one rule associated with the first node is to be associated also with the second node;

based on the identified percolation criterion, associate the at least one rule with the second node such that the at least one rule defines a second factor for determining whether the document is to also be classified into the second node;

access the document for natural language processing; and

classify the document according to the logical hierarchy by determining whether the document is to be classified into at least one of the second node and the first node based on the at least one rule associated with both the first node and the second node.

12. The system of claim 11, wherein the percolation criterion is a first percolation criterion, and the processor is further configured to:

identify a second percolation criterion associated with a third node in the logical hierarchy that is a parent node to the second node, said percolation criterion indicating that the at least one rule associated with the first node is to be associated also with the third node; and

based on the identified second percolation criterion, associate the at least one rule with the third node such that the at least one rule defines a third factor for determining whether the document is to also be classified into the third node.

13. The system of claim 11, wherein the processor is further configured to:

determine that a second percolation criterion is not associated with a third node in the logical hierarchy that is a parent node to the second node; and

based on said determination, limit the at least one rule to be associated only to the first node and the second node.

14. The system of claim 11, wherein classifying the document according to the logical hierarchy comprises analyzing the document utilizing the at least one rule to determine if the document is to be classified into the second node.

15. The system of claim 14, wherein classifying the document according to the logical hierarchy further comprises:

determining that the document is to be classified into the second node based at least in part on the at least one rule; and

based on said determination, analyzing the document utilizing the at least one rule to determine if the document is to be classified into one or more child nodes of the second node, the one or more child nodes including the first node.

16. The system of claim 11, wherein the processor is further configured to:

receive an input to identify which documents in the plurality of documents are to be classified into the first node;

in response to said input, analyze the plurality of documents by the natural language model, using the at least one rule, to determine which documents are to be classified into the first node; and

cause display in a display interface of the documents that are determined to be classified into the first node.

17. The system of claim 16, wherein the processor is further configured to cause display of documents that are determined to be classified into the first node, wherein the determination is made at least in part by the document triggering the at least one rule.

18. A non-transitory computer readable medium comprising instructions that, when executed by a processor, cause the processor to perform operations comprising:

accessing, at a first node in a logical hierarchy configured to guide classification of a plurality of documents by the natural language model, at least one rule associated with the first node, said at least one rule defining a first factor for determining whether a document among the plurality of documents is to be classified into the first node;

identifying a percolation criterion associated with a second node in the logical hierarchy that is a parent node to the first node, said percolation criterion indicating that the at least one rule associated with the first node is to be associated also with the second node;

based on the identified percolation criterion, associating the at least one rule with the second node such that the at least one rule defines a second factor for determining whether the document is to also be classified into the second node;

accessing the document for natural language processing; and

classifying the document according to the logical hierarchy by determining whether the document is to be classified into at least one of the second node and the first node based on the at least one rule associated with both the first node and the second node.

19. The computer readable medium of claim 18, wherein the percolation criterion is a first percolation criterion, and the operations further comprise:

identifying a second percolation criterion associated with a third node in the logical hierarchy that is a parent node to the second node, said percolation criterion indicating that the at least one rule associated with the first node is to be associated also with the third node; and

based on the identified second percolation criterion, associating the at least one rule with the third node such that the at least one rule defines a third factor for determining whether the document is to also be classified into the third node.

20. The computer readable medium of claim 18, wherein classifying the document according to the logical hierarchy comprises:

analyzing the document utilizing the at least one rule to determine if the document is to be classified into the second node;

determining that the document is to be classified into the second node based at least in part on the at least one rule; and

based on said determination, analyzing the document utilizing the at least one rule to determine if the document is to be classified into one or more child nodes of the second node, the one or more child nodes including the first node.