Semantic map generation from natural-language-text documents
AssigneeDigital Asset Capital, Inc.
InventorEdward Hunter
Life of this patent
Estimating the term…
Abstract
A computer-implemented process includes obtaining a natural-language-text document comprising a first and second clause and determining first and second embedding sequences based on n-grams of the first and second clauses. The process includes generating data model objects based on the embedding sequences and determining an association between the first data model object and the second data model object based on a shared parameter of the first and second clauses. The process includes receiving a query including the first category and the first n-gram and causing a presentation of a visualization of data model objects that includes shapes based on the data model objects and a third shape based on the association between the first data model object and the second data model object.
Claims38
A tangible, non-transitory, machine-readable medium storing instructions that when executed by one or more processors effectuate operations comprising:
- obtaining, with a computer system, a natural-language-text document comprising a first clause and a second clause;
- determining, with the computer system, a first embedding sequence based on n-grams of the first clause and a second embedding sequence based on n-grams of the second clause;
- generating, with the computer system, a first data model object and a second data model object based on the first embedding sequence and the second embedding sequence, wherein:
- the first data model object comprises a first field populated with a first n-gram of the first clause and a second field populated with a first category;
- and
- the second data model object comprises third field populated with a second n-gram of the second clause and a fourth field populated with a second category;
- determining, with the computer system, an association between the first data model object and the second data model object based on a shared parameter of the first and second clauses;
- receiving, with the computer system, a query comprising the first category and the first n-gram;
- causing, with the computer system, a presentation of a visualization of the first and second data model objects in response to receiving the query, the visualization comprising:
- a first shape comprising the first n-gram and the first category;
- a second shape comprising the second n-gram and the second category;
- and
- a third shape comprising a first end that is connected to the first shape and a second end that is connected to the second shape based on the association between the first data model object and the second data model object;
- and
- storing, with the computer system, the first data model object and the second data model object in memory.
A method comprising:
- obtaining, with a computer system, a natural-language-text document comprising a first clause and a second clause;
- determining, with the computer system, a first embedding sequence based on n-grams of the first clause and a second embedding sequence based on n-grams of the second clause;
- generating, with the computer system, a first data model object and a second data model object based on the first embedding sequence and the second embedding sequence, wherein:
- the first data model object comprises a first field populated with a first n-gram of the first clause and a second field populated with a first category;
- and
- the second data model object comprises third field populated with a second n-gram of the second clause and a fourth field populated with a second category;
- determining, with the computer system, an association between the first data model object and the second data model object based on a shared parameter of the first and second clauses;
- receiving, with the computer system, a query comprising the first category and the first n-gram;
- causing, with the computer system, a presentation of a visualization of the first and second data model objects in response to receiving the query, the visualization comprising:
- a first shape comprising the first n-gram and the first category
- a second shape comprising the second n-gram and the second category;
- and
- a third shape comprising a first end that is connected to the first shape and a second end that is connected to the second shape based on the association between the first data model object and the second data model object;
- and
- storing, with the computer system, the first data model object and the second data model object in memory.
Dependencies are inferred from the claim text.