Automated ontology system

• US 10,628,743 B1
• Filed: 01/24/2019
• Issued: 04/21/2020
• Est. Priority Date: 01/24/2019
• Status: Active Grant

• Current Assignee: Through Sensing LLC
• Original Assignee: Andrew R. Kalukin
• Inventors: Kalukin, Andrew R.
• Primary Examiner(s): Albertalli, Brian L
• Application Number:
  US16/256,089
• Time in Patent Office: 453 Days
• CPC Class Codes: G06F 16/367 Ontology
• CPC Class Codes: G06F 16/90332 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 40/211 Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
• CPC Class Codes: G06F 40/237 Lexical tools
• CPC Class Codes: G06F 40/289 Phrasal analysis, e.g. finite state techniques or chunking
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/01 Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
• CPC Class Codes: G06N 5/013 Automatic theorem proving
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G06N 5/025 Extracting rules from data
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/225 Feedback of the input speech

Abstract

A robotic system for processing input, such as text data provided through a messaging system, spoken language data provided through a microphone, or any other such input data, which may function to process the input so as to be able to respond or reply to a user based on comprehension of the input sentences. An automated theorem prover (ATP) may operate as an underlying framework for the AI system that understands and responds to spoken or written statements translated into a proper format. An ATP formatter may be used to translate natural language processing (NLP) output from a NLP syntactical sentence parser into the proper format, such that the ATP system may be able to generate and populate an ontology from the NLP output. User queries may be mapped to this ontology in order to facilitate comprehension. If desired, the system may automatically populate the ontology through Internet searching.

Claims

1. A method for automatically generating a response to a user query, comprising:

obtaining a first set of rules defining a knowledge domain;

obtaining a data file of ingestible information;

with a processor, generating a tree structure based on the ingestible information, and converting the tree structure into at least one tree structure first-order logic axiom;

storing a plurality of first-order logic axioms in a memory, the plurality of first-order logic axioms comprising the at least one tree structure first-order logic axiom;

with the processor, parsing further extrinsic information and formulating and testing one or more conjectures based on the further extrinsic information, the one or more conjectures comprising a first-order logic axiom, wherein formulating and testing the one or more conjectures comprises generating, based on a combination of at least two of the plurality of first-order logic axioms, at least one intermediate theorem, and generating, based on the plurality of first-order logic axioms and the at least one intermediate theorem, a conjecture test result, the conjecture test result being an indication that a conjecture in the one or more conjectures is one of true or false;

when the conjecture in the one or more conjectures is true, storing the conjecture in the memory in the form of an ontology;

when the conjecture is false, formulating and testing one or more additional conjectures;

receiving, from an input system, the user query;

with the processor, mapping the user query to the ontology; and

generating, based on the user query and the ontology, the response to the user query.

2. The method of claim 1, wherein the data file of ingestible information comprises text data;

and

wherein the tree structure is a natural language processing (NLP) tree structure representing at least one phrase and comprising a plurality of pairings of syntactic categories and words in the text data.

3. The method of claim 2, wherein the step of converting the tree structure into at least one tree structure first-order logic axiom comprises:

identifying a verb in the tree structure, and automatically creating a theorem based on the verb;

identifying a subject in the tree structure, and adding it to an object list;

identifying a predicate in the tree structure, and adding it to the object list; and

providing the theorem as a tree structure first-order logic axiom of the form “

verb(object list).”

4. The method of claim 2, wherein the step of converting the tree structure into at least one tree structure first-order logic axiom comprises:

identifying a verb in the tree structure, and automatically creating a theorem based on the verb;

identifying a subject in the tree structure, and adding it as an object in an object and feature list;

identifying a predicate in the tree structure, and adding it as an object in the object and feature list;

identifying at least one modifier of at least one of the subject or predicate, and adding the at least one modifier as a feature in the object and feature list in association with the at least one of the subject or predicate; and

providing the theorem as a tree structure first-order logic axiom of the form “

verb(object and feature list),”

wherein the object and feature list comprises at least one object-feature pairing.

5. The method of claim 1, wherein the data file of ingestible information comprises at least one of audio data, image data, or video data.

6. The method of claim 1, wherein the method further comprises mapping, with the processor, the speech data to text data;

and

wherein the tree structure is a natural language processing (NLP) tree structure comprising a plurality of pairings of syntactic categories and words in the text data.

7. The method of claim 1, wherein the user query is added to the data file of ingestible information.

8. The method of claim 1, wherein the method further comprises:

conducting, with the processor and via a network connection, at least one automated web search, and retrieving content from at least one web page; and

adding the content to the data file of ingestible information.

9. An automated ontology system configured to automatically generate a response to a user query, the system comprising an input system, an output system, a processor, and a memory, wherein the system is configured to perform the steps of:

obtaining a first set of rules defining a knowledge domain;

obtaining a data file of ingestible information;

with the processor, generating a tree structure based on the ingestible information, and converting the tree structure into at least one tree structure first-order logic axiom;

storing a plurality of first-order logic axioms in the memory, the plurality of first-order logic axioms comprising the at least one tree structure first-order logic axiom;

with the processor, parsing further extrinsic information and formulating and testing one or more conjectures based on the further extrinsic information, the one or more conjectures comprising a first-order logic axiom, wherein formulating and testing the one or more conjectures comprises generating, based on a combination of at least two of the plurality of first-order logic axioms, at least one intermediate theorem, and generating, based on the plurality of first-order logic axioms and the at least one intermediate theorem, a conjecture test result, the conjecture test result being an indication that a conjecture in the one or more conjectures is one of true or false;

when the conjecture in the one or more conjectures is true, storing the conjecture in the memory in the form of an ontology;

when the conjecture is false, formulating and testing one or more additional conjectures;

receiving, from the input system, the user query;

with the processor, mapping the user query to the ontology; and

generating, based on the user query and the ontology, the response to the user query, and outputting the response to the user query via the output system.

10. The system of claim 9, wherein the data file of ingestible information comprises text data;

and

wherein the tree structure is a natural language processing (NLP) tree structure representing at least one phrase and comprising a plurality of pairings of syntactic categories and words in the text data.

11. The system of claim 10, wherein the step of converting the tree structure into at least one tree structure first-order logic axiom comprises:

identifying a verb in the tree structure, and automatically creating a theorem based on the verb;

identifying a subject in the tree structure, and adding it to an object list;

identifying a predicate in the tree structure, and adding it to the object list; and

providing the theorem as a tree structure first-order logic axiom of the form “

verb(object list).”

12. The system of claim 10, wherein the step of converting the tree structure into at least one tree structure first-order logic axiom comprises:

identifying a verb in the tree structure, and automatically creating a theorem based on the verb;

identifying a subject in the tree structure, and adding it as an object in an object and feature list;

identifying a predicate in the tree structure, and adding it as an object in the object and feature list;

identifying at least one modifier of at least one of the subject or predicate, and adding the at least one modifier as a feature in the object and feature list in association with the at least one of the subject or predicate; and

providing the theorem as a tree structure first-order logic axiom of the form “

verb(object and feature list),”

wherein the object and feature list comprises at least one object-feature pairing.

13. The system of claim 9, wherein the data file of ingestible information comprises at least one of audio data, image data, or video data.

14. The system of claim 9, wherein the system is further configured to map, with the processor, the speech data to text data;

and

wherein the tree structure is a natural language processing (NLP) tree structure comprising a plurality of pairings of syntactic categories and words in the text data.

15. The system of claim 9, wherein the system is further configured to add the user query to the data file of ingestible information.

16. The system of claim 9, wherein the system is further configured to:

conduct, with the processor and via a network connection, at least one automated web search, and retrieve content from at least one web page; and

adding the content to the data file of ingestible information.

17. A non-transitory computer-readable medium comprising computer program code that, when executed, causes a computer system comprising an input system, an output system, a processor, and a memory to perform the steps of:

obtaining a first set of rules defining a knowledge domain;

obtaining a data file of ingestible information;

with the processor, generating a tree structure based on the ingestible information, and converting the tree structure into at least one tree structure first-order logic axiom;

storing a plurality of first-order logic axioms in the memory, the plurality of first-order logic axioms comprising the at least one tree structure first-order logic axiom;

with the processor, parsing further extrinsic information and formulating and testing one or more conjectures based on the further extrinsic information, wherein formulating and testing the one or more conjectures comprises generating, based on a combination of at least two of the plurality of first-order logic axioms, at least one intermediate theorem, and generating, based on the plurality of first-order logic axioms and the at least one intermediate theorem, a conjecture test result, the conjecture test result being an indication that a conjecture in the one or more conjectures is one of true or false;

when the conjecture in the one or more conjectures is true, storing the conjecture in the memory in the form of an ontology;

when the conjecture is false, formulating and testing one or more additional conjectures;

receiving, from the input system, the user query;

with the processor, mapping the user query to the ontology; and

generating, based on the user query and the ontology, the response to the user query, and outputting the response to the user query via the output system.

18. The non-transitory computer-readable medium of claim 17, wherein the data file of ingestible information comprises text data;

and

wherein the tree structure is a natural language processing (NLP) tree structure representing at least one phrase and comprising a plurality of pairings of syntactic categories and words in the text data.

19. The non-transitory computer-readable medium of claim 18, wherein the step of converting the tree structure into at least one tree structure first-order logic axiom comprises:

identifying a verb in the tree structure, and automatically creating a theorem based on the verb;

identifying a subject in the tree structure, and adding it to an object list;

identifying a predicate in the tree structure, and adding it to the object list; and

providing the theorem as a tree structure first-order logic axiom of the form “

verb(object list).”

20. The non-transitory computer-readable medium of claim 18, wherein the step of converting the tree structure into at least one tree structure first-order logic axiom comprises:

identifying a verb in the tree structure, and automatically creating a theorem based on the verb;

identifying a subject in the tree structure, and adding it as an object in an object and feature list;

identifying a predicate in the tree structure, and adding it as an object in the object and feature list;

identifying at least one modifier of at least one of the subject or predicate, and adding the at least one modifier as a feature in the object and feature list in association with the at least one of the subject or predicate; and

providing the theorem as a tree structure first-order logic axiom of the form “

verb(object and feature list),”

wherein the object and feature list comprises at least one object-feature pairing.