System and methods for accessing web pages using natural language
AssigneeGlace Holding LLC
InventorsJohn G. Gorman, John W. R. Gorman
Life of this patent
Estimating the term…
Abstract
A system for accessing a webpage through natural language communication with a computer comprises a registry database configured to store at least one table the table configured to store associations between a concept in a text string and all natural language variations of the text string and associated concept identifiers, the concept in the text string an variations thereof relating to an object; and a parser interfaced with the database and configured to receive the text string or a variation thereof, the parser further configured to identify the concept represented by the text string or variation thereof and associate the identified concept with the object.
Claims27
A system for accessing an object through natural language communication with a computer, the system comprising:
- at least one hardware processor;
- a registry database configured to store at least one table, the table configured to store associations between a concept in a text string and all natural language variations of the text string and associated concept identifiers, wherein the concept in the text string and variations thereof relate to an object;
- and
- a parser interfaced with the database, wherein the parser, when executed by the at least one hardware processor,
- receives the text string or a variation thereof,
- identifies the concept represented by the text string or variation thereof,
- associates the identified concept with the object,
- breaks the text string, or variation thereof, into separate elements of language, the elements including lexical words, grammaticals, and punctuation,
- forms a corresponding ordered string of elementary concepts with unique identifiers associated with the text string,
- identifies sub-concepts pairs comprising two elementary concepts by searching the table for legal joins between component concepts, and
- repeats the following steps until the concept representing the text string is identified and formed:
- forming the sub-concepts by joining component concepts as specified by the table, and
- identifying higher order sub-concepts from the sub-concepts and component concepts by searching the table for legal joins between the component concepts and sub-concepts.
A system for accessing a webpage through natural language communication with a computer, the system comprising:
- at least one hardware processor;
- a portal configured to receive a text string;
- a registry database configured to store at least one table, the table configured to store associations between a concept in a text string and all natural language variations of the text string and associated concept identifiers, wherein the concept in the text string and variations thereof relate to a web address;
- and
- a parser interfaced with the database, wherein the parser, when executed by the at least one hardware processor,
- receives the text string from the portal,
- identifies the concept represented by the text string or variation thereof,
- associates the identified concept with the web address,
- breaks the text string into separate elements of language, the elements including lexical words, grammaticals, and punctuation,
- forms a corresponding ordered string of elementary concepts with unique identifiers associated with the text string,
- identifies sub-concepts pairs comprising two elementary concepts by searching the table for legal joins between component concepts, and
- repeats the following steps until the concept representing the text string is identified and formed:
- forming sub-concepts by joining component concepts as specified by the table, and
- identifying higher order sub-concepts from the sub-concepts and component concepts by searching the table for legal joins between the component concepts and sub-concepts.
A system for accessing a webpage through natural language communication with a computer, the system comprising:
- a database configured to store at least one table, the table configured to store associations between a concept in a text string and all natural language variations of the text string and associated concept identifiers, wherein the concept in the text string and variations thereof relate to a web address;
- a short tag processor configured to generate a short tag that is representative of the text string and associate the short tag with the concept;
- a responder configured to return the short tag;
- and
- a parser interfaced with the database, wherein the parser, when executed by the at least one hardware processor,
- receives the text string or a variation thereof,
- identifies the concept represented by the text string or variation thereof,
- associates the identified concept with the web address,
- breaks the text string, or variation thereof, into separate elements of language, the elements including lexical words, grammaticals, and punctuation,
- forms a corresponding ordered string of elementary concepts with unique identifiers associated with the text string,
- identifies sub-concepts pairs comprising two elementary concepts by searching the table for legal joins between component concepts, and
- repeats the following steps until the concept representing the text string is identified and formed:
- forming sub-concepts by joining component concepts as specified by the table, and
- identifying higher order sub-concepts from the sub-concepts and component concepts by searching the table for legal joins between the component concepts and sub-concepts.
Dependencies are inferred from the claim text.