SYSTEMS AND METHODS TO DETERMINE AND UTILIZE CONCEPTUAL RELATEDNESS BETWEEN NATURAL LANGUAGE SOURCES

• US 20160232160A1
• Filed: 11/25/2015
• Published: 08/11/2016
• Est. Priority Date: 11/26/2014
• Status: Active Grant

• Current Assignee: Vettd, Inc.
• Original Assignee: Vobis, Inc.
• Inventors: Buhrmann, Andrew, Buhrmann, Michael, Smith, Jesse, Keselman, Yakov, Dane, Kurtis Peter, Shokoufandeh, Ali
• Granted Patent: US 11,003,671 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24522 Translation of natural language queries to structured queries
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/3347 using vector based model
• CPC Class Codes: G06F 16/367 Ontology
• CPC Class Codes: G06F 40/14 Tree-structured documents parsing G06F40/205; validation G06F40/226
• CPC Class Codes: G06Q 10/1053 Employment or hiring

Abstract

A microprocessor executable method and system for determining the semantic relatedness and meaning between at least two natural language sources is described in a prescribed context. Portions of natural languages are vectorized and mathematically processed to express relatedness as a calculated metric. The metric is associable to the natural language sources to graphically present the level of relatedness between at least two natural language sources. The metric may be re-determined with algorithms designed to compare the natural language sources with a knowledge data bank so the calculated metric can be ascertained with a higher level of certainty.

Claims

1. A microprocessor executable method to ascertain relatedness between information sources, the microprocessor executable method comprising:

converting natural language of a first information source to a first concept vector;

converting natural language of a second information source to a second concept vector;

determining a similarity between the first concept vector and the second concept vector; and

calculating a metric of the similarity.

2. The microprocessor executable method of claim 1, wherein determining the similarity comprises vector mathematical treatments applied to the first concept vector and the second concept vector.

3. The microprocessor executable method of claim 1, wherein calculating the metric of the similarity includes expressing the metric as at least one of a conceptual relevance score, a conceptually weighted score, a word pool, a first heat map associable with the first information source, a second heat map associable with the second information source, and a graphic representation signifying the evidence of relatedness between the first information source and the second information source.

4. The microprocessor executable method of claim 1, wherein at least the first information source is a ré

sumé

.

6. The microprocessor executable method of claim 4, wherein determining the similarity comprises vector mathematical treatments applied to the first concept vector and the second concept vector.

7. The microprocessor executable method of claim 4, wherein calculating the metric of the similarity includes expressing the metric as at least one of a conceptual relevance score, a conceptually weighted score, a word pool, a first heat map associable with the first information source, a second heat map associable with the second information source, and a graphic representation signifying the evidence of relatedness between the first information source and the second information source.

8. The microprocessor executable method of claim 4, wherein converting the natural language of the first information source includes the content of the natural language within at least one of a classified segment, a plurality of classified segments, an ontological classification, and at least a portion of a spoken communication.

9. The microprocessor executable method of claim 4, wherein converting the natural language of the second information source includes the content of the natural language within at least one of a classified segment, a plurality of classified segments, an ontological classification, and at least a portion of a spoken communication.

11. The microprocessor executable method of claim 9, wherein calculating the metric of the at least one similarity includes expressing the metric as at least one of a conceptual relevance score, a conceptually weighted score, a word pool, a first heat map associable with the first information source, a second heat map the plurality of second order concept vectors that is associable with at least a portion of the concept corpus, and a graphic representation signifying the evidence of relatedness between the first information source and the concept corpus.

12. The microprocessor executable method of claim 9, wherein calculating the metric of the at least one similarity includes expressing the metric as a set of qualification values.

14. The microprocessor executable method of claim 12, wherein notifying the user to modify the first concept vector includes overlaying a text statement near the locus.

16. The microprocessor executable system of claim 14, wherein the storage medium includes a Non-SQL configuration.

20. The microprocessor executable system of claim 14, wherein the at least one similarity is ranked by the metric.

5. A microprocessor executable method to ascertain relatedness between information sources, the microprocessor executable method comprising:

converting natural language of a first information source to a first concept vector;

converting natural language of a second information source to at least one second concept vector;

ontologically comparing the at least one second concept vector with a concept knowledge database;

producing a second order concept vector from the ontologically compared at least one second concept vector;

determining a similarity between the first concept vector and the second order concept vector; and

calculating a metric of the similarity.

10. A microprocessor executable method to ascertain relatedness between information sources, the microprocessor executable method comprising:

partitioning natural language of a first information source into a plurality of information segments;

ontologically comparing the plurality of information segments with a concept knowledge database;

producing a plurality of second order concept vectors from the ontologically compared plurality of information segments;

determining at least one similarity between the plurality of second order concept vectors and a concept corpus; and

calculating a metric of the at least one similarity.

13. A microprocessor executable method to guide a user to modify an information source, the microprocessor executable method comprising:

converting natural language of a first information source to a first concept vector;

obtaining a plurality of second concept vectors from a concept knowledge database;

determining at least one similarity between the first concept vector and the plurality of second order concept vectors;

identifying a locus in the first information source having significant relevance of the first concept vector with the at least one similarity; and

notifying the user to modify the first concept vector at the locus within the first information source.

15. A microprocessor executable system configured to ascertain relatedness between information sources, the microprocessor executable method comprising:

a storage medium configured to store natural language of a first information source and a second information source; and

a processor configured to execute instructions to convert the natural language of the first information source to a first concept vector and to convert the natural language of the second information source to a second concept vector; and

determining at least one similarity between the first concept vector and the second concept vector; and

calculating a metric of the at least one similarity.

17. The microprocessor executable system 14, wherein the similarity includes at least one of a text relevance, a conceptual relevance, a first order concept map, a second order concept map, and a contextually weighted relevance.

18. The microprocessor executable system 14, wherein the metric includes at least one of a conceptual relevance score, a conceptually weighted score, a word pool, a first heat map associable with the first information source, a second heat map the plurality of second order concept vectors that is associable with at least a portion of the concept corpus, and a graphic representation signifying the evidence of relatedness between the first information source and the concept corpus.

19. The microprocessor executable system 14, wherein the at least one similarity includes an overlay expressible on a webpage.